Functions in R
The syntax for writing functions in R is very straightforward and there is little room for errors. If you know some of the most common functions in R like the apply family, you can create and customize a lot of functions.
In the first part of this function , we tell R that this is going to be a function be writing “function”. Then we add the arguments to the function viz., “x” and “y” within parentheses. Next, within the curly brackets we write the actual mechanics of the function. In this case we want to add x and y together. Let’s see how this function works.
x <- 2
y <- 4
functiona(x,y)
[1] 6What is tidyevals
Above we showed a very simple example of a function in R. But you can do so much more with these functions. For example, if you have a template for ggplot, you can encase it within a function and then just change the name of the variables to get fresh outputs, without the need for copying and pasting.
Rows: 32
Columns: 11
$ mpg <dbl> 21.0, 21.0, 22.8, 21.4, 18.7, 18.1, 14.3, 24.4, 22.8, 1…
$ cyl <dbl> 6, 6, 4, 6, 8, 6, 8, 4, 4, 6, 6, 8, 8, 8, 8, 8, 8, 4, 4…
$ disp <dbl> 160.0, 160.0, 108.0, 258.0, 360.0, 225.0, 360.0, 146.7,…
$ hp <dbl> 110, 110, 93, 110, 175, 105, 245, 62, 95, 123, 123, 180…
$ drat <dbl> 3.90, 3.90, 3.85, 3.08, 3.15, 2.76, 3.21, 3.69, 3.92, 3…
$ wt <dbl> 2.620, 2.875, 2.320, 3.215, 3.440, 3.460, 3.570, 3.190,…
$ qsec <dbl> 16.46, 17.02, 18.61, 19.44, 17.02, 20.22, 15.84, 20.00,…
$ vs <dbl> 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1…
$ am <dbl> 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1…
$ gear <dbl> 4, 4, 4, 3, 3, 3, 3, 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 4, 4…
$ carb <dbl> 4, 4, 1, 1, 2, 1, 4, 2, 2, 4, 4, 3, 3, 3, 4, 4, 4, 1, 2…## Suppose you create a graph for cylinder count and mpg of cars using this data set
mtcars |>
ggplot(aes(x = disp, y = mpg)) +
geom_point(aes(color = wt, size = wt)) +
facet_wrap(~ cyl)+
theme(legend.position = "NULL")
Now that we have created this graph suppose we want to plot the relationship between horsepower and mpg in a similar fashion. One option is to write the samy syntax once more and then change the y value to “hp”. Or you can write a function to automate the process.
# So given the information you already have, the function will look like this
graph<- function(independent, dependent){
mtcars |>
ggplot(aes(x = independent, y = dependent)) +
geom_point(aes(color = wt, size = wt)) +
facet_wrap(~ cyl)+
theme(legend.position = "NULL")
}
# lets see if this works
graph(independent = disp,dependent = hp)
# We get an error out of this
Why didn’t this work? The answer lies in tidyeval. You see when you are calling variables within a function they should be within ellipsis or quotation marks. Since our function does not know where to look for the variables disp and mpg, it is not able to compute the function. So now we when we call the variables dependent and independent within the function syntax, we will wrap them in double curly brackets. This will ensure that the variables are treated as variables and the function should work.
graph<- function(independent, dependent){
mtcars |>
ggplot(aes(x = {{independent}}, y = {{dependent}})) +
geom_point(aes(color = wt, size = wt)) +
facet_wrap(~ cyl)+
theme(legend.position = "NULL")
}
graph(independent = disp,dependent = hp)
So this is what tidyeval is in the simplest terms. If you are writing functions in R where different variables involved, you must wrap them in double curly braces, or your function will not work.
Other examples
We have seen one way in which writing functions can help you declutter your code. Now lets talk about how we can chain up multi-step data analysis processes using functions. Suppose you have a dirty data set that you want to wrangle regularly and pre-process without having to write the same functions over and over again. For example, whenever I get a new data-set I will typically run “janitor::clean_names” followed by “janitor::remove_empty_rows” and “janitor:: remove_empty_cols” to clean up the mess a little bit.
cleanitup<- function(dataset){
library(janitor)
dataset<<- dataset |> clean_names() |>
janitor::remove_empty()
}
cleanitup(mtcars)
glimpse(mtcars)
Rows: 32
Columns: 11
$ mpg <dbl> 21.0, 21.0, 22.8, 21.4, 18.7, 18.1, 14.3, 24.4, 22.8, 1…
$ cyl <dbl> 6, 6, 4, 6, 8, 6, 8, 4, 4, 6, 6, 8, 8, 8, 8, 8, 8, 4, 4…
$ disp <dbl> 160.0, 160.0, 108.0, 258.0, 360.0, 225.0, 360.0, 146.7,…
$ hp <dbl> 110, 110, 93, 110, 175, 105, 245, 62, 95, 123, 123, 180…
$ drat <dbl> 3.90, 3.90, 3.85, 3.08, 3.15, 2.76, 3.21, 3.69, 3.92, 3…
$ wt <dbl> 2.620, 2.875, 2.320, 3.215, 3.440, 3.460, 3.570, 3.190,…
$ qsec <dbl> 16.46, 17.02, 18.61, 19.44, 17.02, 20.22, 15.84, 20.00,…
$ vs <dbl> 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1…
$ am <dbl> 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1…
$ gear <dbl> 4, 4, 4, 3, 3, 3, 3, 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 4, 4…
$ carb <dbl> 4, 4, 1, 1, 2, 1, 4, 2, 2, 4, 4, 3, 3, 3, 4, 4, 4, 1, 2…# Since we saved the modified mtcars to dataset, now we can access it outside the function as well
glimpse(dataset)
Rows: 32
Columns: 11
$ mpg <dbl> 21.0, 21.0, 22.8, 21.4, 18.7, 18.1, 14.3, 24.4, 22.8, 1…
$ cyl <dbl> 6, 6, 4, 6, 8, 6, 8, 4, 4, 6, 6, 8, 8, 8, 8, 8, 8, 4, 4…
$ disp <dbl> 160.0, 160.0, 108.0, 258.0, 360.0, 225.0, 360.0, 146.7,…
$ hp <dbl> 110, 110, 93, 110, 175, 105, 245, 62, 95, 123, 123, 180…
$ drat <dbl> 3.90, 3.90, 3.85, 3.08, 3.15, 2.76, 3.21, 3.69, 3.92, 3…
$ wt <dbl> 2.620, 2.875, 2.320, 3.215, 3.440, 3.460, 3.570, 3.190,…
$ qsec <dbl> 16.46, 17.02, 18.61, 19.44, 17.02, 20.22, 15.84, 20.00,…
$ vs <dbl> 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1…
$ am <dbl> 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1…
$ gear <dbl> 4, 4, 4, 3, 3, 3, 3, 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 4, 4…
$ carb <dbl> 4, 4, 1, 1, 2, 1, 4, 2, 2, 4, 4, 3, 3, 3, 4, 4, 4, 1, 2…If you want to save a object created within your function to the global environment, you can use the “<<-” operator rather than the common “<-”. Writing functions in R is not very difficult or complicated once you have grasped the core concepts behind it.