How To Make Density Plots with ggplot2 in R?

In this tutorial, we will see how to make density plots with ggplot2 in R. We will start with an example of making a basic density plot with ggplot2 and see multiple examples of making the density plots better.

Let us load tidyverse, a suite R packages from RStudio and also set the default theme to theme_bw() for all the plots we make.

library(tidyverse)
theme_set(theme_bw(base_size=16))

To make density plots we will use 2019 Stack Overflow survey data. The results from the survey is processed and is available at datavizpyr.com‘s github page.

stackoverflow_file <- "https://raw.githubusercontent.com/datavizpyr/data/master/SO_data_2019/StackOverflow_survey_filtered_subsampled_2019.csv"

survey_results <- read_csv(stackoverflow_file)

The survey data is a fantastic data source for understanding data science/developers work life with many interesting variables. In this tutorial, we will mainly use distribution of developers salary in US to make density plots with ggplot2.

## # A tibble: 5 x 4
##   CompTotal Gender Manager YearsCode
##       <dbl> <chr>  <chr>   <chr>    
## 1    180000 Man    IC      25       
## 2     55000 Man    IC      5        
## 3     77000 Man    IC      6        
## 4     67017 Man    IC      4        
## 5     90000 Man    IC      6

Let us make the most basic density plot with ggplot2 in R. With ggplot2, we can make density plot using geom_density() function. We specify x-axis aesthetics, the variable we want to make density plot to ggplot’s aes() function and add geom_density() as another layer to make density plot. In this example code we have below, we also specify x axis label and show how to save the plot as jpg file using ggsave() function.

survey_results %>% 
  ggplot(aes(x=CompTotal)) + 
  geom_density() +
  labs(x="Developer Salary")
ggsave("simple_density_plot_ggplot2_R.jpg")

The basic density plot looks like this with mean salary around 100K, but with a few outlier developers with very high salaries.

Density plot with ggplot2 with fill color

Let us improve the density plot with adding a color manually to fill the density curve. We can fill the density plot with a color using fill argument. We can use fill with in aes() function or within geom_density() function. In this example, we are manually filling the plot with a specific color with fill=”dodgerblue” inside geom_density() function.

survey_results %>% 
  ggplot(aes(x=CompTotal)) + 
  geom_density(fill="dodgerblue")  +
  labs(x="Developer Salary")
ggsave("simple_density_plot_with_color_ggplot2_R.jpg")

The density plot with fill color look like this.

Our density plot is skewed because outlier developer salary. A better way to handle the skewness due to outliers is to change the scale the of the data. Here we can make the density plot with log scale on x-axis.

Density plot with ggplot2 in log scale

In ggplot2, we can change the x-axis scale to log scale by adding scale_x_log10() function. scale_x_log10() function will show the developer salary on a logarithmic scale, instead of the original scale.

survey_results %>% 
  ggplot(aes(x=CompTotal)) + 
  geom_density(fill="dodgerblue")  +
  scale_x_log10()+
  labs(x="Developer Salary")

Our density plot in R using logarithmic scale on x-axis looks much better now.

Density plot with ggplot2 with transparent color fill

We can also change the transparency level of the fill color by setting alpha values ranging from 0 to 1 inside geom_density() function.

survey_results %>% 
  ggplot(aes(x=CompTotal)) + 
  geom_density(fill="dodgerblue", alpha=0.5)  +
  scale_x_log10()+
  labs(x="Developer Salary")

In this example, we have set the transparency level to 50% with alpha=0.5.