Data visualization with R and ggplot2

Data visualization with ggplot2 in detail

Data visualization with R and ggplot2, also known as the Grammar of Graphics, is a free, open-source, and user-friendly visualization package widely utilized in the R programming language. Created by Hadley Wickham, it is one of the most powerful tools for data visualization.

Key Layers of ggplot2

The ggplot2 package operates on several layers, which include:

1. Data: The dataset used for visualization.
2. Aesthetics: Mapping data attributes to visual properties such as x-axis, y-axis, color, fill, size, labels, alpha, shape, line width, and line type.
3. Geometric Objects: How data is represented visually, such as points, lines, histograms, bars, or boxplots.
4. Facets: Splitting data into subsets displayed in separate panels using rows or columns.
5. Statistics: Applying transformations like binning, smoothing, or descriptive summaries.
6. Coordinates: Mapping data points to specific spaces (e.g., Cartesian, fixed, polar) and adjusting limits.
7. Themes: Customizing non-data elements like font size, background, and color.

Dataset Used: mtcars

The mtcars dataset contains fuel consumption and 10 other automobile design and performance attributes for 32 cars. It comes pre-installed with the R environment.

Viewing the First Few Records

# Print the first 6 records of the dataset
head(mtcars)

Output:

mpg	cyl	disp	hp	drat	wt	qsec	vs	am	gear	carb
Mazda RX4	21.0	6	160	110	3.90	2.620	16.46	0	1	4	4
Mazda RX4 Wag	21.0	6	160	110	3.90	2.875	17.02	0	1	4	4
Datsun 710	22.8	4	108	93	3.85	2.320	18.61	1	1	4	1
Hornet 4 Drive	21.4	6	258	110	3.08	3.215	19.44	1	0	3	1
Hornet Sportabout	18.7	8	360	175	3.15	3.440	17.02	0	0	3	2
Valiant	18.1	6	225	105	2.76	3.460	20.22	1	0	3	1

Summary Statistics of mtcars

# Load dplyr package and get a summary of the dataset
library(dplyr)

# Summary of the dataset
summary(mtcars)

Output:

Variable	Min	1st Quartile	Median	Mean	3rd Quartile	Max
mpg	10.4	15.43	19.20	20.09	22.80	33.90
cyl	4.0	4.0	6.0	6.19	8.0	8.0
disp	71.1	120.8	196.3	230.7	326.0	472.0
hp	52.0	96.5	123.0	146.7	180.0	335.0
drat	2.76	3.08	3.70	3.60	3.92	4.93
wt	1.51	2.58	3.32	3.22	3.61	5.42
qsec	14.5	16.89	17.71	17.85	18.90	22.90
vs	0.0	0.0	0.0	0.44	1.0	1.0
am	0.0	0.0	0.0	0.41	1.0	1.0
gear	3.0	3.0	4.0	3.69	4.0	5.0
carb	1.0	2.0	2.0	2.81	4.0	8.0

Visualizing Data with ggplot2

Data Layer: The data layer specifies the dataset to visualize.

# Load ggplot2 and define the data layer
library(ggplot2)

ggplot(data = mtcars) +
  labs(title = "Visualization of MTCars Data")

Output:

Aesthetic Layer: Mapping data to visual attributes such as axes, color, or shape.

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

Output:

Geometric Layer: Adding geometric shapes to display the data.

# Plot data using points
plot1 <- ggplot(data = mtcars, aes(x = hp, y = mpg, col = disp)) +
  geom_point() +
  labs(title = "Horsepower vs Miles per Gallon", x = "Horsepower", y = "Miles per Gallon")

Output:

Faceting: Create separate plots for subsets of data.

# Facet by transmission type
facet_plot <- ggplot(data = mtcars, aes(x = hp, y = mpg, shape = factor(cyl))) +
geom_point()
facet_grid()}

Output:

Statistics Layer: The statistics layer in ggplot2 allows you to transform your data by applying methods like binning, smoothing, or descriptive statistics.

# Scatter plot with a regression line
ggplot(data = mtcars, aes(x = hp, y = mpg)) +
  geom_point() +
  stat_smooth(method = lm, col = "blue") +
  labs(title = "Relationship Between Horsepower and Miles per Gallon")

Output:

Coordinates Layer: In this layer, data coordinates are mapped to the plot’s visual space. Adjustments to axes, zooming, and proportional scaling of the plot can also be made here.

# Scatter plot with controlled axis limits
ggplot(data = mtcars, aes(x = wt, y = mpg)) +
  geom_point() +
  stat_smooth(method = lm, col = "green") +
  scale_y_continuous("Miles per Gallon", limits = c(5, 35), expand = c(0, 0)) +
  scale_x_continuous("Weight", limits = c(1, 6), expand = c(0, 0)) +
  coord_equal() +
  labs(title = "Effect of Weight on Fuel Efficiency")

Output:

Using coord_cartesian() to Zoom In

# Zoom into specific x-axis and y-axis ranges
ggplot(data = mtcars, aes(x = wt, y = hp, col = as.factor(am))) +
  geom_point() +
  geom_smooth() +
  coord_cartesian(xlim = c(3, 5), ylim = c(100, 300)) +
  labs(title = "Zoomed View: Horsepower vs Weight",
       x = "Weight",
       y = "Horsepower",
       color = "Transmission")

Output:

Theme Layer: The theme layer in ggplot2 allows fine control over display elements like background color, font size, and overall styling.

Example 1: Customizing the Background with element_rect()

ggplot(data = mtcars, aes(x = hp, y = mpg)) +
geom_point() +
facet_grid(. ~ cyl) +
theme(plot.background = element_rect(fill = "lightgray", colour = "black")) +
labs(title = "Background Customization: Horsepower vs MPG")

Output:

Example 2: Using theme_gray()

ggplot(data = mtcars, aes(x = hp, y = mpg)) +
geom_point() +
facet_grid(am ~ cyl) +
theme_gray() +
labs(title = "Default Theme: Horsepower and MPG Facets")

Output:

Contour Plot for the mtcars Dataset: Create a density contour plot to visualize the relationship between two continuous variables.

# 2D density contour plot
ggplot(mtcars, aes(x = wt, y = mpg)) +
  stat_density_2d(aes(fill = ..level..), geom = "polygon", color = "black") +
  scale_fill_viridis_c() +
  labs(title = "2D Density Contour: Weight vs MPG",
       x = "Weight",
       y = "Miles per Gallon",
       fill = "Density Levels") +
  theme_minimal()

Output:

Creating a Panel of Plots: Create multiple plots and arrange them in a grid for side-by-side visualization.

library(gridExtra)

# Histograms for selected variables
hist_plot_mpg <- ggplot(mtcars, aes(x = mpg)) +
  geom_histogram(binwidth = 2, fill = "steelblue", color = "black") +
  labs(title = "Miles per Gallon Distribution", x = "MPG", y = "Frequency")

hist_plot_disp <- ggplot(mtcars, aes(x = disp)) +
  geom_histogram(binwidth = 50, fill = "darkred", color = "black") +
  labs(title = "Displacement Distribution", x = "Displacement", y = "Frequency")

hist_plot_hp <- ggplot(mtcars, aes(x = hp)) +
  geom_histogram(binwidth = 20, fill = "forestgreen", color = "black") +
  labs(title = "Horsepower Distribution", x = "Horsepower", y = "Frequency")

hist_plot_drat <- ggplot(mtcars, aes(x = drat)) +
  geom_histogram(binwidth = 0.5, fill = "orange", color = "black") +
  labs(title = "Drat Distribution", x = "Drat", y = "Frequency")

# Arrange plots in a 2x2 grid
grid.arrange(hist_plot_mpg, hist_plot_disp, hist_plot_hp, hist_plot_drat, ncol = 2)

Output:

Saving and Extracting Plots

To save plots as image files or reuse them later:

# Create a plot
plot <- ggplot(data = mtcars, aes(x = hp, y = mpg)) +
  geom_point() +
  labs(title = "Horsepower vs MPG")

# Save the plot as PNG
ggsave("horsepower_vs_mpg.png", plot)

# Save the plot as PDF
ggsave("horsepower_vs_mpg.pdf", plot)

# Extract the plot for reuse
extracted_plot <- plot
plot

Output: