Skip to content

Vectors

A vector is a fundamental data structure in R and can be considered as a basic building block for storing and manipulating data. In R, vectors are one-dimensional datasets that consist of elements of the same data type. Unlike matrices and data frames, which are two-dimensional, vectors represent a single sequence of values without any row or column structure.

mycharacter_vector <- c("un", "deux", "trois", "quatre", "cinq", "six")
mynumeric_vector <- c(1, 2, 3, 4, 5, 6)
myboolean_vector <- c(TRUE, FALSE, FALSE, TRUE, FALSE, TRUE)

Above three variables are character, numeric and logical vectors, respectively.

And you can see their content just by typing and entering their names: 

mycharacter_vector
## [1] "un"     "deux"   "trois"  "quatre" "cinq"   "six"   
mynumeric_vector
## [1] 1 2 3 4 5 6
myboolean_vector
## [1]  TRUE FALSE FALSE  TRUE FALSE  TRUE

Learn more about vectors with the chapter 5.1 of Garrett Grolemund’s Hands-on Programing with R’s book. The R guide by Nathaniel D. Phillips shows how to create vectors in section 5.2 and how to manipulate them in sections 6 and 7.

If you want to test your knowledge, don’t hesitate to do the exercises proposed by Nathaniel D. Phillips where you can retrieve the solutions in sections 18.2 to 18.4.

Attributes

An attribute is a piece of information that you can attach to an atomic vector (or any R object). The attribute will not affect any of the values in the object, and it will not appear when you display your object. You can think of an attribute as "metadata"; it is just a convenient place to put information associated with an object. R will normally ignore this metadata, but some R functions will check for specific attributes. These functions may use the attributes to do special things with the data.

Attributes can be set and accessed individually with attr().

For instance:

sd <- 1:10
attr(sd, "comment") <- "This is a bad name for a vector because sd is also a built-in function name in R"

sd
## [1]  1  2  3  4  5  6  7  8  9 10

attr(sd, "comment")
## [1] "This is a bad name for a vector because sd is also a built-in function name in R"

The three most important predefined attributes are:

1. Names

A character vector giving each element a name. See the section on names in the Advanced R by Hadley Wickham. Names are undoubtedly attributes that you will use extensively. Thus, let's give quickly to a vector a names attribute:

results <- c(1.2, 3.4, 8.02)

results
## [1] 1.20 3.40 8.02

names(results) <- c("Rep1", "Rep2", "Rep3")

results
## Rep1 Rep2 Rep2 
## 1.20 3.40 8.02

## Or using a short form with setNames()
results <- setNames(c(1.2, 3.4, 8.02), c("Rep1", "Rep2", "Rep3"))
results
## Rep1 Rep2 Rep2 
## 1.20 3.40 8.02

You can also give names to a vector when you create it:

other_results <- c(Rep1 = 5.6, Rep2 = 0.9, Rep3 = 5.7, Rep4 = 7.65)

other_results
## Rep1 Rep2 Rep3 Rep4 
## 5.60 0.90 5.70 7.65

Finally, you can give names to a vector by creating a modified copy of a vector:

other_results <- setNames(other_results, c("replicat1", "replicat2", "replicat3", "replicat4"))

other_results
## replicat1 replicat2 replicat3 replicat4 
##      5.60      0.90      5.70      7.65

2. Dimensions

Used to turn vectors into matrices and arrays. This will be covered later.

3. Class

Used to implement the S3 object system; this will be covered later too.

Exercises on vectors

Exercises to manipulate vectors with operators and functions may be found here.

Factors

A factor is an R object that represents an R vector in a compact way.

To build the factor of a vector, R, behind the scene, (i) determines the distinct values that are present in the vector, which are called levels (ii) assigns an unique integer value to each level and (iii) stores for each value of the initial vector, the corresponding level, expressed with its assigned integer value.

Example-1

Let’s start from the vector

pidigits <- c(3, 1, 4, 1, 5, 9, 3)

and generate its factor pidigits_factor by

pidigits_factor <- factor(pidigits)

You may now compare the two objects:

pidigits
## [1] 3 1 4 1 5 9 3
pidigits_factor
## [1] 3 1 4 1 5 9 3
## Levels: 1 3 4 5 9

At first glance, the two objects are similar, except that the pidigits_factor has now levels values stored along the initial vector, as a metadata. In addition, if you look at the structure of the pidigits_factor object, you will see that it has been built as explained above:

str(pidigits_factor)
## Factor w/ 5 levels "1","3","4","5",..: 2 1 3 1 4 5 2

Thus, the value 2 has been assigned to the level "3" of Pi and comes at first in the initial vector, the value 1 has been assigned to the level "1" of Pi and comes at the second and fourth positions of the original vector, respectively, the value 3 has been assigned to the level "4" of Pi, etc.

Example-2

You have 4 mice and you differentiate males and females in a vector:

mice <- c("female", "male","male", "female")

And you build the factor object mice_factor with:

mice_factor <- factor(mice)

Then:

mice_factor
## [1] female male   male   female
## Levels: female male

str(mice_factor)
## Factor w/ 2 levels "female","male": 1 2 2 1

As you see, the levels are "female" and "male", internally encoded by the value 1 and 2, respectively. Thus the internal sequence 1 2 2 1 returns the vector

[1] female male   male   female

To learn how to manipulate factors, go through sections 5.5.2 of Garrett Grolemund’s guide and R’s factor of GeeksforGeeks page: https://hbctraining.github.io/Intro-to-R-flipped/lessons/02_introR-syntax-and-data-structures.html.

Last, note that a factor is an object with two attributes, the attribute class and the attribute levels. Thus:

class(mice_factor)
## [1] "factor"
levels(mice_factor)
## [1] "female" "male"

## Or:
attr(mice_factor, "class")
## [1] "factor"
attr(mice_factor, "levels")
## [1] "female" "male"

Note that the class attribute "factor" tells R that mice_factor behave differently from regular vectors.

Variable Manipulation

Now that you have mastered the differences between variable’s types and structures, you need to manipulate them.

Following some references for data manipulation: