Aesthetics best practices

Introduction to Data Visualization with ggplot2

Rick Scavetta

Founder, Scavetta Academy

Which aesthetics?

• Use your creative know-how, and
• Follow some clear guidelines
• Jacques Bertin
  • The Semiology of Graphics, 1967
• William Cleveland
  • The Elements of Graphing Data, 1985
  • Visualizing Data, 1993

Form follows function

Form follows function

Function

Primary:

• Accurate and efficient representations

Secondary:

• Visually appealing, beautiful plots

Guiding principles

Never:

• Misrepresent or obscure data
• Confuse viewers with complexity

Always:

• Consider the audience and purpose of every plot

Extracting information from Data

The best choices for aesthetics

• Efficient
  • Provides a faster overview than numeric summaries
• Accurate
  • Minimizes information loss

Aesthetics - continuous variables

ggplot(iris, aes(x = Sepal.Length,
                 y = Sepal.Width,
                 color = Species)) + 
  geom_point()

Aesthetics - continuous variables

ggplot(iris, aes(color = Sepal.Length,
                 y = Sepal.Width,
                 x = Species)) + 
  geom_point()

Three iris scatter plots

Three iris scatter plots, unaligned y-axes

Single faceted plot, common y-axis

Aesthetics - categorical variables

ggplot(iris, aes(x = Sepal.Length,
                 y = Sepal.Width, 
                 col = Species)) + 
  geom_point()

Aesthetics - categorical variables

ggplot(iris, aes(x = Sepal.Length, 
                 y = Sepal.Width, 
                 col = Species)) + 
  geom_point(position = "jitter", 
             alpha = 0.5)

Now it's your turn

Introduction to Data Visualization with ggplot2