Functions in R

Day 1 - Introduction to Data Analysis with R

Selina Baldauf

Freie Universität Berlin - Theoretical Ecology

March 14, 2025

Functions in R

Functions make multiple operations available under one command.


Functions:

  • Have a name
  • Have (0, 1, or any number of) arguments as input
  • They calculate something using the arguments
  • They have a return value (the output)
  • Can have side effects (like plotting)


General structure of a function call: function_name ( argument = value )

The mean function

function_name ( argument = value )


mean ( x = c(1,5,6) )

mean(x = c(1,5,6)) # or short: mean(c(1,5,6))
#> [1] 4
  • Arguments can also be variables
  • The output of a function can be stored in a variable
values <- c(1,5,6)
result <- mean(x = values)
result
#> [1] 4

The mean function

But what does the mean function do? What are the arguments that I can use?

  Call the function help using ?

?mean

Function arguments

  • Arguments are the input to a function
  • Functions can provide default values for some arguments
  • Default values for arguments are indicated in the function help

Function arguments

What happened here?   

# NA is a missing value
values <- c(1, 5, 6, NA)
mean(x = values)
#> [1] NA

  na.rm argument is FALSE by default.

Set it to TRUE if you want to calculate the mean despite missing values:

mean(x = values, na.rm = TRUE)
#> [1] 4


Arguments with default values are optional, arguments without default values are not! 

mean()
#> Error in mean.default(): argument "x" is missing, with no default

Function arguments

Argument matching can be achieved by position or by name

mean(x, trim = 0, na.rm = FALSE, ...)
values <- c(1, 5, 6, NA)

These calls to mean all are the same:

mean(values, , TRUE) # by position
mean(x = values, na.rm = TRUE) # by name
mean(values, na.rm = TRUE) # a mix of both

Function arguments

Argument matching can be achieved by position or by name

Named arguments are (generally) preferred

  • Easier to remember
  • Easier to read
  • Some functions have a lot of arguments
mean(x = values, na.rm = TRUE) # by name

However, it is common to match the first argument by position (especially when the first argument is the data)

mean(values, na.rm = TRUE) # a mix of both

Where do functions come from?

Base R functions: built into R

mean() # calculate mean
seq()  # generate a sequence of values

From additional packages

  • Packages must be installed first
  • Call a function from a package using packageName::functionName()
  • Load the package with library(packageName) and then use the function


# use read_csv function from readr package
readr::read_csv()

# or use library()
library(readr)
read_csv()

Custom functions:

Write your own functions and then use them in the code

# custom function that prints input in a sentence
my_function <- function(x) {
  return(x + 5)
}
my_function(5)

Summary

R basics - Functions

Summary

  • Functions take input in the form of (named) arguments, calculate something and return a result

  • Functions are called by their name, followed by parentheses:
    functionName(argument1 = value, argument2 = value, ...)

  • Functions from additional packages can be called in two ways:

    • packageName::functionName()
    • first load the package with library(packageName) then call the function anywhere in the script with functionName()

  • Call ?functionName to open the help of a function