Plotting Overview and Refresher

Resources:

• Data Carpentry - Data Analysis and Visualisation for Ecologists
• R for Data Science - Visualisation

Refresher

A more in depth intro to ggplot2 is covered in Data Carpentry - Data Analysis and Visualisation for Ecologists. This is a brief overview/refresher.

Setup

library(tidyverse)
library(palmerpenguins)

# brings 'penguins' dataset into name space
attach(penguins)
# brings 'penguins_raw' dataset into name space
attach(penguins_raw)

Basics of Plotting with {ggplot2}

The basic premise of ggplot is that there are 3 main components to a plot:

• the data
• a mapping of the data to aesthetics of the plot e.g. x or y axes
  • this can be done for all layers of a plot at once by including it within the aes() within the ggplot() function
  • or to specific layers by using aes() within a geom function
• a geometric representation of the data e.g points, bars, boxplot

and the plots are built in layers

# general structure for a ggplot
ggplot(data = <DATA>, mapping = aes(<MAPPINGS>)) + <GEOM_FUNCTION>()

ggplot(
  data = penguins,
  mapping = aes(x = body_mass_g,
                y = flipper_length_mm)
) +
  geom_point()

Warning: Removed 2 rows containing missing values or values outside the scale range
(`geom_point()`).

And we can modify the properties of layer to a specific value

ggplot(
  data = penguins,
  mapping = aes(x = body_mass_g,
                y = flipper_length_mm)
) +
  geom_point(colour = "blue")

Warning: Removed 2 rows containing missing values or values outside the scale range
(`geom_point()`).

Or the property can be mapped back to the data - when this is the case we use aes()

ggplot(
  data = penguins,
  mapping = aes(x = body_mass_g,
                y = flipper_length_mm)
) +
  geom_point(mapping = aes(colour = island))

Warning: Removed 2 rows containing missing values or values outside the scale range
(`geom_point()`).

Plots can also be built incrementally by assignment

penguin_plot <- ggplot(
  data = penguins,
  mapping = aes(x = body_mass_g,
                y = flipper_length_mm,
                colour = island)
) 

penguin_plot <- penguin_plot +
  geom_point()

penguin_plot

Warning: Removed 2 rows containing missing values or values outside the scale range
(`geom_point()`).