Tidyverse Cheatsheet

All examples use built-in data (data('iris')) so you can copy and paste the examples. This is not an exhaustive manual, but describes the main approaches used in the book.

Installing and loading the entire tidyverse

install.packages('tidyverse') # only needed once
library(tidyverse)

The pipe operator `|>`

The pipe operator is used in the tidyverse to chain together actions described by the verbs. It works by injecting the output of one function as first argument in the next function:

# Pipe operator
# Output of function1 is injected into first argument of position2
function1(input, arg1) |> function2(arg2)

# Explicit nesting
# This performs the exact same computation as the previous example
function2(function1(input, arg1), arg2)

# Slightly more abstract
x |> f(y) == f(x, y)

Core `dplyr` verbs used in this course

`filter()`: select rows (observations) in a tidy dataset

Typically used with a logical expression using comparison operators (e.g. ==, >=, etc., see ‘comparison operators’ in the base R cheatsheet)

# Select all observations of the setosa iris species
iris |>
  filter(Species == 'setosa')

`select()`: select columns (variables) in a tidy dataset

# Select only petal width and petal length variables
iris |>
  select(Petal.Width, Petal.Length)

`group_by()`: create groups based on variables in the data

# Create groups for individual species
iris |> 
  group_by('Species')

`mutate()`: create new columns/variables (based on existing ones)

# Create a new variable of the ratio between petal length and width
iris |>
  mutate(
    petal_ratio = Petal.Length / Petal.Width
  )

`summarise()`: create summaries of existing variables

# Compute the mean and median petal width
iris |>
  summarise(
    mean_petal_width = mean(Petal.Width),
    median_petal_width = median(Petal.Width)
  )