Subsetting data

R has many powerful subset operators and mastering them will allow you to easily perform complex operation on any kind of dataset. Allows you to manipulate data very succinctly.

As the last section for this topic we'll cover:

The three subsetting operators,
The six types of subsetting,
Important difference in subsetting behaviour for different objects
Using subsetting in conjunction with assignment

Subsetting atomic vectors

x <- c(5.4, 6.2, 7.1, 4.8, 7.5)

We can subset this in many ways

1. Using positive integers

x[1]  
x[c(3,1)]  
x[1:4]  
# We can duplicate indices  
x[c(1, 1)]  
# Real numbers are silently truncated to integers  
x[c(2.1, 2.9)]

2. Using negative integers

# skip the first element  
x[-1]  
# skip the first and the fifth  
x[-c(1,5)]

3. Using logical operators

x[c(TRUE, TRUE, FALSE, FALSE)]
# or based on a condition
x[x > 3]
x[which(x >3)]
Also see `which.max()` and `which.min()`

4. with nothing

x[]
# useful when dealing with 2 or higher dimensional objects

5. with zero

x[0]
# helpful for generating test data or creating empty objects

6. Referencing objects by their names

(y <- setNames(x, letters[1:4]))

y[c("d", "c", "a")]

# We can also repeat indices
y[c("a", "a", "a")]

# Names are always matched exactly, not partially
z <- c(abc = 1, def = 2)
z[c("a", "d")]
# <NA> <NA> 
#   NA   NA

Subsetting lists

Subsetting a list works in exactly the same way as subsetting an atomic vector. Subsetting a list with [ will always return a list: [[ and $, as described below, let you pull out the components of the list.

x <- as.list(1:10)

x[1:5]

# What class is this object?

To extract individual elements inside a list, use the [[ operator

# to get element 5

x[[5]]

class(x[[5]])

another_variable <- x[[5]]

Or using their names

names(x) <- letters[1:5]

x$a
x[["a"]]

Subsetting matrices

a <- matrix(1:9, nrow = 3)
colnames(a) <- c("A", "B", "C")
a[1:2, ]
#      A B C
# [1,] 1 4 7
# [2,] 2 5 8
a[c(T, F, T), c("B", "A")]
#      B A
# [1,] 4 1
# [2,] 6 3
a[0, -2]
#      A C

When you extract a single column or row, you get a vector out. This is not always what is wanted, so the drop argument comes in useful

a[1,,drop=FALSE]
a[1,]

Subsetting data frames

df <- data.frame(x = 1:3, y = 3:1, z = letters[1:3])

df[df$x == 2, ]
#   x y z
# 2 2 2 b
df[c(1, 3), ]
#   x y z
# 1 1 3 a
# 3 3 1 c

# There are two ways to select columns from a data frame
# Like a list:
df[c("x", "z")]
#   x z
# 1 1 a
# 2 2 b
# 3 3 c
# Like a matrix
df[, c("x", "z")]
#   x z
# 1 1 a
# 2 2 b
# 3 3 c

# There's an important difference if you select a simple column:
# matrix subsetting simplifies by default, list subsetting does not.
df["x"]
#   x
# 1 1
# 2 2
# 3 3
df[, "x"]
# [1] 1 2 3

If you use the single bracket subset you get a new dataframe out. If you use $ or [[ you get the contents of that column. If the column does not exist, $ will return NULL and [[ will return an error.

v <- "x"
df[[v]]
df$v # this does not work!
df$x # this does
df$not_in_here # NULL
df[["not_in_here"]] # error

Exercises

1. Fix each of the following common data frame subsetting errors:

# Extract cases where cyl is 4
mtcars[mtcars$cyl = 4, ]

# Exclude only rows 1 through 4
mtcars[-1:4, ]

# Return only rows for cylinders less than 5
mtcars[mtcars$cyl <= 5]

# Return only rows for cylinders that are 4 or 6.
mtcars[mtcars$cyl == 4 | 6, ]

2. Why does mtcars[1:20] return a error? How does it differ from the similar mtcars[1:20, ]?

3. R comes with a data set called iris;

How big is this dataset (number of rows and columns)?
Create a new data.frame called small_diamonds that only contains rows 1 through 9 and 19 through 23. You can do this in one or two steps.

4. Given a linear model

set.seed(10)
dat <- data.frame(response=rnorm(30, mean=rep(c(2, 5, 10), each=10), sd=rep(0.1, 30)), factors=rep(LETTERS[1:3], each=10))
mod <- aov(response ~ factors, data = dat)

Extract the residual degrees of freedom.