Data Types

We will be using a variety of objects to operationalize our use of R. Here we’ll discuss briefly the important types of objects and how they should be used.

Basic R

Objects and OOP

R is an Object-Oriented Programming (OOP) language which means it functions best by storing things as “objects”. An object is a data structure that has some attributes and a set of methods that act on those attributes.

Object-oriented languages (often called “high-level” languages) are typically more intuitive, as whenever you want to assign some value(s) to a variable and save those values for later, you can make the variable to an object which will be saved by R in the environment for future use.

Data Types

There are six main data types in R, but we will discuss only the first four.

  1. Characters
  2. Numerics (real or decimal)
  3. Intergers
  4. Logicals
  5. Complex
  6. Raw

When we assign variables, it is important to know what type we are using. We can check the type of an object, and get other useful information about it, with a number of useful commands:

  x <- "abc"
  typeof(x)
## [1] "character"

If we want a more detailed answer we can ask what the structure (str()) of an object is

  str(x)
##  chr "abc"

We see the output is slightly different here, with the function telling us the type of the object (chr) and also the content of the object ("abc").

Characters

A character is any alphanumeric string that begins with a letter.

  x <- "abc"
  y <- "abc123"
  typeof(x)
## [1] "character"
  typeof(y)
## [1] "character"

When referring to numbers in a computer, it’s important to remember that some numbers, π for example, are infinite. However, computers can’t actually hold infinite decimal places of a value, so we have to decide how accurate we want our values. R is required to work with values conforming to a global standard, which works out to a precision of 53 bits (what is a bit?!). This is called “binary64” format and is “double” the standard precision of “binary32”, hence being called double precision

Numerics

A numeric type is any non-integer number.

  z <- 3.14
  typeof(z)
## [1] "double"

The output "double"” here refers to the fact that R automatically stores numeric data types with “double” precision.

Integers

An integer is a non-decimal whole number. Note: In R, the default is to store values as numeric unless explicitly told otherwise. We can see that if we make a new variable with only an integer value R will store it as a numeric type.

  x1 <- 2
  typeof(x1)
## [1] "double"

To force R to store it as an integer we can simply add an L after the value.

  x2 <- 2L
  typeof(x2)
## [1] "integer"

Logicals

Logical types are simply TRUE or FALSE.

  x3 <- TRUE
  typeof(x3)
## [1] "logical"