This Section

You can take an entire class on data visualization. Sadly, the guiding principles and theory of data visualization is beyond the scope of our current class.

So what will you learn in this section?

1. (This lecture) Reminder of the basics of ggplot2 & some new things
2. Changing the gestalt/look/feel/mood of your plot
   • Color palettes
   • Themes
   • Manually changing very specific things
3. Breaking down and adding to plots
   • Make a multi-panel figure from your data
   • Combine separate plots into a single multi-panel figure
   • Add things like lines, text etc.
4. Going further
   • Making scatterplots more readable
   • Combine multiple types of plot into a single, informative figure

What you'll need

Please make sure the following packages are installed:

• tidyverse (this includes dplyr and ggplot2)
• RColorBrewer
• ggpubr
• if you like Wes Anderson movies (The Royal Tennenbaums, Life Aquatic, Isle of Dogs, Darjeeling Limited, Rushmore, The Grand Budapest Hotel, and more!), install the wesanderson package

Remember: do NOT memorize anything! You will always have these slides to refer back to, and Google is your friend. Looking things up is a good thing!

Recap of ggplot2

Recap of ggplot2

• The gg in ggplot2 stands for "the grammar of graphics"
• ggplot2 has the following structure:

ggplot(things that impact the entire plot) + 
  geom_something(things that impact just the something)

Plotting with ggplot2

ggplot2 has the following structure:

ggplot(things that impact the entire plot) +
  geom_something(things that impact just the something)

Things like:

• data.frame used for plotting
• defining your x & y axes

Layer 1: ggplot

What happens if you only input your data and what you want your x- and y-axes to be?

ggplot(data = empire, aes(x = mass, y = height))

Layer 2: geom_

ggplot2 has the following structure:

ggplot(things that impact the entire plot) + 
  geom_something(things that impact just the something)

geom_ typically means shape. What shapes do you want to use to represent your data in the plot?

• geom_histogram -- histogram
• geom_density -- distributions
• geom_violin -- distributions
• geom_point -- scatter plot
• geom_col -- bar plot

These are functions that are searchable in the help pages!

Plotting with ggplot2

The functions ggplot() and geom_() can take on different aesthetics as an argument, using aes().

Aesthetics are how you control what you want your plot to look like; how can you make it pretty? Examples:

• Which variables are the x- and y- axes?
• color (should you color the plot by some variable?)
• fill (very similar to color, should you fill the plot in somehow; used for bar graphs and boxplots)
• shape (do you want groups to have different shaped points?)
• size (how big should plotted data be?)

When to use aesthetics

ggplot(data = empire,
       aes(x = mass,
           y = height)) +
  geom_point()

When to use aesthetics

ggplot(data = empire,
       aes(x = mass,
           y = height)) +
  geom_point(color = "cornflowerblue")

When to use aesthetics

ggplot(data = empire,
       aes(x = mass,
           y = height)) +
  geom_point(aes(color = species))

Mix & Match

ggplot(data = empire,
       aes(x = mass,
           y = height)) +
  geom_point(color = "purple",
             aes(shape = species),
             size = 3)

Layer 3: labs

ggplot(data = empire,
       aes(x = mass,
           y = height)) +
  geom_point(aes(color = species,
                 shape = species)) +
  labs(x = "Weight in kg",
       y = "Height in cm",
       title = "Size of Star Wars Characters")

Overlapping Data

• alpha = is an argument that controls transparency (1 = opaque, 0 = transparent)
• Without setting the alpha parameter...

ggplot(data = empire,
       aes(x = height)) +
  geom_density(aes(fill = sex)) + 
  labs(x = "Height in cm", 
       y = "Probability Density", 
       title = "Height Distributions")

Overlapping Data

• alpha = is an argument that controls transparency (1 = opaque, 0 = transparent)
• With setting the alpha parameter...

ggplot(data = empire,
       aes(x = height)) +
  geom_density(aes(fill = sex),
               alpha = .5) +
  labs(x = "Height in cm", 
       y = "Probability Density", 
       title = "Height Distributions")

Re-organizing Factors

• In R, the levels of each factor are based on alphabetical order
• This means when you go to plot, your factors will show up in the same order.
• For example:
  • geom_bar() uses count as default for the y-axis
  • That means, it gives you the same information as table()

table(empire$species)

## 
##          Droid          Human        Wookiee Yoda's species 
##              2              6              1              1

Re-organizing Factors

ggplot(data = empire,
       aes(x = species)) +
  geom_bar(aes(fill = species))

Re-organizing Factors

Let's say we want the order (from left to right) to be Human, Yoda's species, Wookiee, Droid.

• To do this, we need to re-order the species factor like so:

empire$species <- factor(empire$species,
                         levels = c("Human",
                                    "Yoda's species",
                                    "Wookiee",
                                    "Droid"))

Now, if we simply go the tables of counts again, it should match our new ordering:

table(empire$species)

## 
##          Human Yoda's species        Wookiee          Droid 
##              6              1              1              2

Re-organizing Factors

ggplot(data = empire,
       aes(x = species)) +
  geom_bar(aes(fill = species))

Next time...

• How to assign certain colors to certain factor levels (e.g., make "Droid" blue and "Yoda's species" maroon)
• Color palettes
• Themes
• Changing the background of your plot