Data types

Jeff Stevens

2023-02-03

Review

Mental model of literate programming

Data types

Data types

Formats of individual elements

• Double

• Integer

• Character

• Logical

Check data types with typeof()

Mental model of data types

Numeric data

Doubles

Floating-point numbers with decimals

# assign value 7.2 to object a
(a <- 7.2) # remember, wrapping in parentheses prints to console

[1] 7.2

typeof(a)

[1] "double"

Numeric data

Integers

Numbers without decimals

(b <- 7)

[1] 7

typeof(b)

[1] "double"

Doubles can have 0 as decimal

Numeric data

Integers

Numbers without decimals (specified with L)

(c <- 7L)

[1] 7

typeof(c)

[1] "integer"

Character data

Must be surrounded by ""

(d <- "Hello, world")

[1] "Hello, world"

(e <- "7")

[1] "7"

typeof(e)

[1] "character"

Let’s explore!

• Type b, c, and e into the console separately. What do you see?
• Add b + c.
• Add b + e.

Logical

Tests whether conditional statement is TRUE or FALSE (notice always all upper case)

Logical operators: >, >=, <, <=, ==, !=, %in%

a

[1] 7.2

a > 5

[1] TRUE

d

[1] "Hello, world"

(mytest <- d == "Good-bye, world")

[1] FALSE

typeof(mytest)

[1] "logical"

Logical

The logical operator for equals is ==

Note

We use

• == for logical equals
• <- for assigning objects
• = for assigning function argument values to argument names

Logical

%in% operator: “is contained in”

(subjects <- c("01", "02", "03", "04", "05"))

[1] "01" "02" "03" "04" "05"

"03" %in% subjects

[1] TRUE

"06" %in% subjects

[1] FALSE

Test “is NOT contained in” with ! before test string

!"06" %in% subjects

[1] TRUE

Augmented data types

Core data types with special attributes

• Factors

• Dates

Factors

Augmented integers with ‘levels’

(i <- factor("married", levels = c("single", "married", "widowed")))

[1] married
Levels: single married widowed

typeof(i)

[1] "integer"

Note

Use class() to view augmented data type

class(i)

[1] "factor"

Dates

Augmented numerics based on number of days since 1970-01-01

(j <- as.Date("1970-01-01"))

[1] "1970-01-01"

typeof(j)

[1] "double"

class(j)

[1] "Date"

Note

Make sure to wrap dates in ""

Dates

You can do math on dates

(k <- as.Date("2023-02-03"))

[1] "2023-02-03"

k-j

Time difference of 19391 days

Check data types

• Check with typeof()/class()

• Check in RStudio

• Use is.<type>() functions:
  is.logical(), is.numeric(), is.character()

is.character(7)

[1] FALSE

is.character("7")

[1] TRUE

Converting between data types (coercion)

Use as.<type>() functions:

as.logical(), as.numeric(), as.character()

e

[1] "7"

typeof(e)

[1] "character"

(l <- as.numeric(e))

[1] 7

typeof(l)

[1] "double"

Converting between data types (coercion)

(m <- "TRUE")

[1] "TRUE"

typeof(m)

[1] "character"

(n <- as.logical(m))

[1] TRUE

typeof(n)

[1] "logical"

Converting between data types (coercion)

Factors to numerics is tricky

(o <- factor(0, levels = c("1", "0")))

[1] 0
Levels: 1 0

as.numeric(o)

[1] 2

First coerce to character

as.character(o)

[1] "0"

as.numeric(as.character(o))

[1] 0

Special values

NA represents missing values

• Each data type has its own type of NA

• Check with is.na()

NaN means “not a number” (undefined)

• 0 / 0 = NaN

Inf and -Inf represent infinity and negative infinity

• 1 / 0 = Inf

• -1 / 0 = -Inf

Mental model of data types

Let’s code!

Data types coding [Rmd]