ggplot2 in R

We devise visualizations on mtcars dataset which includes 32 car brands and 11 attributes using ggplot2 layers.

Data Layer:

ggplot2 in R the data Layer we define the source of the information to be visualize, let’s use the mtcars dataset in the ggplot2 package.

R

library(ggplot2)
library(dplyr)
 
ggplot(data = mtcars) +
  labs(title = "MTCars Data Plot")

Output:

ggplot2 in R

Aesthetic Layer:

ggplot2 in R Here we will display and map dataset into certain aesthetics.

R

# Aesthetic Layer
ggplot(data = mtcars, aes(x = hp, y = mpg, col = disp))+
 labs(title = "MTCars Data Plot")

Output:

ggplot2 in R

Geometric layer:

ggplot2 in R geometric layer control the essential elements, see how our data being displayed using point, line, histogram, bar, boxplot.

R

# Geometric layer
ggplot(data = mtcars, aes(x = hp, y = mpg, col = disp)) +
  geom_point() +
  labs(title = "Miles per Gallon vs Horsepower",
       x = "Horsepower",
       y = "Miles per Gallon")

Output:

Data visualization with R and ggplot2

Geometric layer: Adding Size, color, and shape and then plotting the Histogram plot

R

# Adding size
ggplot(data = mtcars, aes(x = hp, y = mpg, size = disp)) +
  geom_point() +
  labs(title = "Miles per Gallon vs Horsepower",
       x = "Horsepower",
       y = "Miles per Gallon")
 
# Adding shape and color
ggplot(data = mtcars, aes(x = hp, y = mpg, col = factor(cyl),
       shape = factor(am))) +geom_point() +
  labs(title = "Miles per Gallon vs Horsepower",
       x = "Horsepower",
       y = "Miles per Gallon")
 
# Histogram plot
ggplot(data = mtcars, aes(x = hp)) +
  geom_histogram(binwidth = 5) +
  labs(title = "Histogram of Horsepower",
       x = "Horsepower",
       y = "Count")

Output:

Data visualization with R and ggplot2

Facet Layer:

ggplot2 in R facet layer is used to split the data up into subsets of the entire dataset and it allows the subsets to be visualized on the same plot. Here we separate rows according to transmission type and Separate columns according to cylinders.

R

# Facet Layer
# Separate rows according to transmission type
p <- ggplot(data = mtcars, aes(x = hp, y = mpg, shape = factor(cyl))) + geom_point()
 
p + facet_grid(am ~ .) +
  labs(title = "Miles per Gallon vs Horsepower",
       x = "Horsepower",
       y = "Miles per Gallon")
 
# Separate columns according to cylinders
p <- ggplot(data = mtcars, aes(x = hp, y = mpg, shape = factor(cyl))) + geom_point()
 
p + facet_grid(. ~ cyl) +
  labs(title = "Miles per Gallon vs Horsepower",
       x = "Horsepower",
       y = "Miles per Gallon")

Output:

Data visualization with R and ggplot2Data visualization with R and ggplot2

Data visualization with R and ggplot2

Statistics layer

ggplot2 in R this layer, we transform our data using binning, smoothing, descriptive, intermediate

R

ggplot(data = mtcars, aes(x = hp, y = mpg)) +
  geom_point() +
  stat_smooth(method = lm, col = "red") +
  labs(title = "Miles per Gallon vs Horsepower")

Output:

Data visualization with R and ggplot2

Coordinates layer:

ggplot2 in R these layers, data coordinates are mapped together to the mentioned plane of the graphic and we adjust the axis and changes the spacing of displayed data with Control plot dimensions.

R

ggplot(data = mtcars, aes(x = wt, y = mpg)) +
  geom_point() +
  stat_smooth(method = lm, col = "red") +
  scale_y_continuous("Miles per Gallon", limits = c(2, 35), expand = c(0, 0)) +
  scale_x_continuous("Weight", limits = c(0, 25), expand = c(0, 0)) +
  coord_equal() +
  labs(title = "Miles per Gallon vs Weight",
       x = "Weight",
       y = "Miles per Gallon")

Output:

Data visualization with R and ggplot2

Coord_cartesian() to proper zoom in:

R

# Add coord_cartesian() to proper zoom in
ggplot(data = mtcars, aes(x = wt, y = hp, col = am)) +
                        geom_point() + geom_smooth() +
                        coord_cartesian(xlim = c(3, 6))

Output:

Data visualization with R and ggplot2

Theme Layer:

ggplot2 in R layer controls the finer points of display like the font size and background color properties.

Example 1: Theme layer – element_rect() function

R

ggplot(data = mtcars, aes(x = hp, y = mpg)) +
  geom_point() +
  facet_grid(. ~ cyl) +
  theme(plot.background = element_rect(fill = "blue", colour = "gray")) +
  labs(title = "Miles per Gallon vs Horsepower")

Output:

Data visualization with R and ggplot2

Example 2:

R

ggplot(data = mtcars, aes(x = hp, y = mpg)) +
        geom_point() + facet_grid(am ~ cyl) + 
        theme_gray()+
 labs(title = "Miles per Gallon vs Horsepower")

Output:

Data visualization with R and ggplot2

ggplot2 in R provides various types of visualizations. More parameters can be used included in the package as the package gives greater control over the visualizations of data. Many packages can integrate with the ggplot2 package to make the visualizations interactive and animated.

Contour plot for the mtcars dataset

R

# Install and load required packages
install.packages("ggplot2")
library(ggplot2)
 
# Create a 2D density contour plot for the mtcars dataset
ggplot(mtcars, aes(x = wt, y = mpg)) +
  stat_density_2d(aes(fill = ..level..), geom = "polygon", color = "white") +
  scale_fill_viridis_c() + 
  labs(title = "2D Density Contour Plot of mtcars Dataset",
       x = "Weight (wt)",
       y = "Miles per Gallon (mpg)",
       fill = "Density") +
  theme_minimal()

Output:

Data visualization with R and ggplot2

In ggplot2 in R stat_density_2d to generate the 2D density contour plot. The aesthetics x and y specify the variables on the x-axis and y-axis, respectively. The fill aesthetic is set to ..level.. to map fill color to density levels.

Creating a panel of different plots

R

library(ggplot2)
library(gridExtra)
 
# Selecting specific columns from mtcars dataset
selected_cols <- c("mpg", "disp", "hp", "drat")
selected_data <- mtcars[, selected_cols]
 
# Create histograms for individual variables
hist_plot_mpg <- ggplot(selected_data, aes(x = mpg)) +
  geom_histogram(binwidth = 2, fill = "blue", color = "white") +
  labs(title = "Histogram: Miles per Gallon", x = "Miles per Gallon", y = "Frequency")
 
hist_plot_disp <- ggplot(selected_data, aes(x = disp)) +
  geom_histogram(binwidth = 50, fill = "red", color = "white") +
  labs(title = "Histogram: Displacement", x = "Displacement", y = "Frequency")
 
hist_plot_hp <- ggplot(selected_data, aes(x = hp)) +
  geom_histogram(binwidth = 20, fill = "green", color = "white") +
  labs(title = "Histogram: Horsepower", x = "Horsepower", y = "Frequency")
 
hist_plot_drat <- ggplot(selected_data, aes(x = drat)) +
  geom_histogram(binwidth = 0.5, fill = "orange", color = "white") +
  labs(title = "Histogram: Drat", x = "Drat", y = "Frequency")
 
# Arrange the plots in a grid
grid.arrange(hist_plot_mpg, hist_plot_disp, hist_plot_hp, hist_plot_drat,
             ncol = 2)

Output:

Data visualization with R and ggplot2

The ggplot2 and gridExtra packages to create histograms for four different variables (“Miles per Gallon,” “Displacement,” “Horsepower,” and “Drat”) from the mtcars dataset.

Each histogram is visually represented in a distinctive color (blue, red, green, and orange) with white borders. The resulting grid of histograms provides a quick visual overview of the distribution of these car-related variables.

Save and extract R plots:

To save and extract plots in R, you can use the ggsave function from the ggplot2 package. Here’s an example of how to save and extract plots:

R

# Create a plot
plot <- ggplot(data = mtcars, aes(x = hp, y = mpg)) +
  geom_point() +
  labs(title = "Miles per Gallon vs Horsepower")
 
# Save the plot as an image file (e.g., PNG)
ggsave("plot.png", plot)
 
# Save the plot as a PDF file
ggsave("plot.pdf", plot)
 
# Extract the plot as a variable for further use
extracted_plot <- plot
plot

Output:

Data visualization with R and ggplot2

In this demonstration, I used ggplot to construct a plot and the ggsave function to save it as a PDF file (plot.pdf) and a PNG image file (plot.png). By including the correct file extension, you can indicate the intended file format.

You may easily give the ggplot object to a variable, as demonstrated with extracted_plot, to extract the plot as a variable for later usage.

Be sure to substitute your unique plot and desired file names for the plot code and file names (plot.png and plot.pdf).

Data Layer:

R

Aesthetic Layer:

R

Geometric layer:

R

R

Facet Layer:

R

Statistics layer

R

Coordinates layer:

R

R

Theme Layer:

R

R

Contour plot for the mtcars dataset

R

Creating a panel of different plots

R

Save and extract R plots:

R

Data visualization with R and ggplot2

Similar Reads

Contact Us