Blog

An Easy Guide to Writing Functions in R

Definition:

What is a function in R? A function in R is similar to a function in calculus because they both take input(s), perform operations on those input(s) and produce an output.

So in calculus a function is something like:

f(x) = 2x + 1; where 2x + 1 is the operation to be performed.

If we input x = 1. We get an output of:

f(1) = 2(1) + 1 = 3

Functions in R have the same fundamental idea. We need:

1. Input(s) also called arguments in R
2. To define the operation that will produce the output called the function

For the R demonstrations in this tutorial, I will be using RMarkdown. Feel free to use base R as you follow along. I will also assume each person utilizing this tutorial has downloaded R studio. Otherwise, this is a link to download R Studio.

Working Environment

In R studio, click on file -> new file -> R Markdown:

General structure of an R function

Before we begin, let’s explore the overall structure of an R function.

The overall structure of an R function has three important components:

1. Name- name the function in a way that specifies what the function does (this is good coding practice). E. g. If we want to write a function that calculates the mean blood pressure, we can name it: Avgbp <-

2. Detail the input (or argument). In R this is done with the function called function. E.g. function (a), where a is the input value for the function.

3. Body of the function which is the code that defines the operations on the input to give the desired output. This is usually written in curly brackets.

So in brief, the general structure of a function should look like this:

Avgbp <- function(a) {

Body

}

Example:

Let’s compute the average systolic blood pressure (sbp) of patients as shown below:

Group 1:

Patient a: sbp = 120

Patient b: sbp = 135

Patient c: sbp = 115

Group 2:

Patient d: sbp = 118

Patient e: sbp = 127

Patient f: sbp = 109

Group 3:

Patient g: sbp = 139

Patient h: sbp = 162

Patient i: sbp = 105

Group 4:

Patient j: sbp = 102

Patient k: sbp = 133

Patient l: sbp = 129

#Average sbp group 1: 
meansbp1 <- sum(120, 135, 115)/3
meansbp1

## [1] 123.3333

#Average sbp group 2: 
meansbp2 <- sum(118, 127, 109)/3
meansbp2

## [1] 118

#Average sbp group 3: 
meansbp3 <- sum(139, 162, 105)/3
meansbp3

## [1] 135.3333

#Average sbp group 4: 
meansbp4 <- sum(102, 133, 129)/3
meansbp4

## [1] 121.3333

To calculate the average systolic blood pressure of the four groups, I had to copy and paste the formular I used for the first group 3 times.

Rule of Thumb

The rule of thumb in R programming is that if you have copied and pasted a formular two times, rather than repeat this a third time: WRITE A FUNCTION!

Let’s Write a function to calculate the average systolic blood pressure

Avgsbp <- function(x) {
n <- length(x)
sum.x <- sum(x)
mean <- sum.x/n
return(mean)
}

Elements of the function above explained

1. Name: We named our function Avgsbp
2. Input: We have only one input for our function which is sbp. Note: Functions can have more than one input
3. Operation 1: We defined the length of our values and placed it in a vector called n
4. Operation 2: We defined the sum.x vector to calculate the sum of the values of x
5. Operation 3: We defined the mean vector to calculate the mean using sum.x and n
6. Output: We used the return function to call our output

Let’s test our function

#Average sbp group 1
Avgsbp(c(120,135,115))

## [1] 123.3333

#Average sbp group 2
Avgsbp(c(118, 127, 109))

## [1] 118

#Average sbp group 3
Avgsbp(c(139, 162, 105))

## [1] 135.3333

Avgsbp(c(102, 133, 129))

## [1] 121.3333