Your First R Instructions

1. Calculations

Arithmetic operators work like a calculator: + adds, - subtracts, * multiplies, / divides, and ^ raises to a power. Parentheses control what happens first.

2 + 3

[1] 5

10 / 2

[1] 5

(4 + 6) * 2

[1] 20

(4 + 6) / 2

[1] 5

Predict the result of each line before running it.

Two more operators are useful with whole numbers. %% gives the remainder after division, and %/% gives the whole-number part of a division. They help with questions such as “is this number even?” or “how many full weeks are in 100 days?”

2 ^ 3        # 2 to the power 3

[1] 8

17 %% 5      # remainder of 17 divided by 5

[1] 2

17 %/% 5     # how many whole 5s fit in 17

[1] 3

100 %/% 7    # full weeks in 100 days

[1] 14

100 %% 7     # days left over

[1] 2

17 %% 5 prints 2 because 17 is three 5s (15) plus 2, and 17 %/% 5 prints the 3.

2. Variables: Give a result a name

R’s assignment symbol <- means “store the value on the right under the name on the left.” It does not mean “the two things are equal forever.”

age <- 54
age                          # Read the stored value

[1] 54

age_next_year <- age + 1
age_next_year

[1] 55

age <- 55                    # Replaces the previous age value

Running age <- 55 makes the name age point at 55 and drops 54, while age_next_year keeps the 55 it was given.

R also accepts a single = for assignment, so this works too:

age = 54
age

[1] 54

It prints 54, the same as with <-. This course uses <- for assignment because it is the common R style and our convention, not because = is wrong. It also lets = keep one job, naming arguments inside a function (as in digits = 2 below), while == compares.

Names are case-sensitive: Age and age are different objects. Prefer descriptive lowercase names with underscores (for example, systolic_bp). Avoid spaces in object names.

3. Text, numbers, and TRUE/FALSE

R handles several fundamental data types:

• numeric: 2, 2.5, or -10.
• character: text enclosed by quotation marks, such as "North".
• logical: TRUE or FALSE, without quotation marks.
• missing (NA): an unknown value; not the same as zero.

weight_kg <- 70.5
clinic <- "Mobile"
is_adult <- TRUE
missing_bmi <- NA

class(weight_kg)

[1] "numeric"

class(clinic)

[1] "character"

class(is_adult)

[1] "logical"

is.na(missing_bmi)

[1] TRUE

The three class() lines print "numeric", "character" and "logical", and is.na(missing_bmi) prints TRUE because that value is unknown. Note that "Mobile" needs quotes but TRUE does not.

A numeric-looking identifier such as "00123" should often remain text. Identifiers are labels, not measurements to average.

4. Integer, double, and the special values

Numbers come in two types. A double is a number that may have decimals; plain numbers you type, such as 5, are doubles. An integer is a whole number, and you make one by adding L, as in 5L. typeof() shows the stored type. For everyday work you rarely need to care, but counts such as length() are integers, and you will see “integer” and “double” in tables and error messages.

typeof(5)

[1] "double"

typeof(5L)

[1] "integer"

typeof(2.5)

[1] "double"

class(5L)

[1] "integer"

R also has special values that are not ordinary numbers:

• Inf and -Inf: infinity, from dividing a nonzero number by zero.
• NaN: “not a number”, from an undefined calculation such as 0 / 0.
• NULL: nothing at all, an empty object. It is not the same as NA, which is a missing value in a place where a value belongs.

1 / 0

[1] Inf

-1 / 0

[1] -Inf

0 / 0

[1] NaN

is.nan(0 / 0)

[1] TRUE

length(NULL)

[1] 0

length(NA)

[1] 1

1 / 0 prints Inf, 0 / 0 prints NaN, and length(NULL) is 0 while length(NA) is 1, because NA still occupies one slot. If a calculation on clinic data returns Inf or NaN, look for a zero in the denominator, such as a height of 0.

5. Functions: Give R a job

A function is a named operation followed by parentheses. Values inside the parentheses are called arguments. Some arguments have names.

round(3.14159, digits = 2)

[1] 3.14

sqrt(81)

[1] 9

sum(2, 5, 8)

[1] 15

round(3.14159, digits = 2) prints 3.14, sqrt(81) prints 9 and sum(2, 5, 8) prints 15. In the first call, digits is the name of the second argument, and 2 is its value.

Type help("round") in your own Console to open the help page for a function. It is shown here, not run, because it opens a help page rather than printing a result.

help("round")

When you see round(3.14159, digits = 2), read it aloud as “round 3.14159 to two digits after the decimal point.”

6. Math functions

R has the usual math functions built in. Each takes a number (or a whole vector of numbers) and returns a number.

x <- 7.58
round(x, 1)      # nearest tenth

[1] 7.6

ceiling(x)       # round up to a whole number

[1] 8

floor(x)         # round down

[1] 7

trunc(-x)        # drop the decimals, toward zero

[1] -7

sqrt(49)

[1] 7

exp(1)           # e to the power 1

[1] 2.718282

log(100)         # natural logarithm

[1] 4.60517

log10(100)       # base-10 logarithm

[1] 2

log2(8)          # base-2 logarithm

[1] 3

factorial(5)     # 5 * 4 * 3 * 2 * 1

[1] 120

sin(0)

[1] 0

cos(0)

[1] 1

ceiling(7.58) is 8 and floor(7.58) is 7. trunc(-7.58) is -7 (toward zero), whereas floor(-7.58) would be -8. log(100) is about 4.6, but log10(100) is exactly 2. sin() and cos() work in radians and are rarely needed in clinic data; log() is the one you will meet most, for skewed measurements.

7. What is an object? typeof(), class(), and attributes

Everything you store in R is an object: a number, a piece of text, a vector, or later a whole table. Two questions describe an object. typeof() says how it is stored inside R, and class() says what kind of thing it is for the functions that use it. Often they agree, but not always (a factor is stored as integers).

An object can also carry extra labels called attributes, such as names. attributes() lists them and attr() reads or sets one.

pulse <- c(a = 70, b = 82)
typeof(pulse)

[1] "double"

class(pulse)

[1] "numeric"

attributes(pulse)

$names
[1] "a" "b"

attr(pulse, "units") <- "beats/min"
attr(pulse, "units")

[1] "beats/min"

attributes(pulse)

$names
[1] "a" "b"

$units
[1] "beats/min"

attributes(pulse) first lists only $names ("a" and "b"); after the attr() line it lists $units too. You rarely write attributes yourself, but this is where a factor keeps its levels and a table keeps its column names.

8. Comparisons and logical values

Comparison operators produce TRUE or FALSE. Use == to compare values; do not confuse it with <- to assign. Other comparisons are >, <, >=, <=, and !=.

age <- 54
age >= 18

[1] TRUE

age == 54

[1] TRUE

age != 54

[1] FALSE

(age > 40) & (age < 65)      # Both conditions must be true

[1] TRUE

(age < 20) | (age > 50)      # Either condition may be true

[1] TRUE

With age equal to 54, the five lines print TRUE, TRUE, FALSE, TRUE and TRUE. The last one is TRUE because the second condition, age > 50, holds even though age < 20 does not.

R uses & for “and” and | for “or” when comparing vectors element by element.

9. Comments and reading errors

Text after a # is for humans; R does not execute it.

# This is a useful explanation for your future self.
bmi <- 27.2
bmi * 2

[1] 54.4

R skips the comment line, stores 27.2 under bmi, and prints 54.4. Typing bmii * 2 instead would give “object ‘bmii’ not found”, because that name was never created.

If R prints “object not found,” check whether you created that object, ran the earlier lines, and used the exact capitalization. If it prints “unexpected symbol,” look for missing quotation marks, commas, and parentheses.

weight_kg <- 72
height_m <- 1.65
bmi <- weight_kg / height_m^2
bmi

[1] 26.44628

bmi <- 72 / 1.65^2
high_bmi <- bmi >= 30
high_bmi

[1] FALSE

id_text <- "00123"
id_number <- 123
class(id_text)

[1] "character"

class(id_number)

[1] "numeric"

id_text

[1] "00123"

id_number

[1] 123

bmi <- 26.4
msg <- tryCatch(bmii * 2, error = function(e) conditionMessage(e))
msg

[1] "object 'bmii' not found"

bmi * 2

[1] 52.8

250 %% 7

[1] 5

250 %/% 7

[1] 35

typeof(250)

[1] "double"

typeof(250L)

[1] "integer"

0 / 0

[1] NaN

1 / 0

[1] Inf

length(NULL)

[1] 0

x <- 3.7
round(x)

[1] 4

ceiling(x)

[1] 4

floor(x)

[1] 3

trunc(x)

[1] 3

sqrt(x)

[1] 1.923538

log10(x)

[1] 0.5682017

factorial(4)

[1] 24

bp <- c(120, 135)
attr(bp, "units") <- "mmHg"
attributes(bp)

$units
[1] "mmHg"