Understanding the cut Function in R with Breaks

The cut function in R is a powerful tool for dividing and categorizing data into specified intervals or bins. In this article, we will delve deeper into how the cut function works, especially when it comes to specifying breaks. We’ll explore some common questions and edge cases that users may encounter.

Setting Up the Environment

Before we dive in, let’s create a sample dataset to work with.

set.seed(111)
data1 <- seq(1,10, by=1)

This code generates a sequence of numbers from 1 to 10 and assigns it to the data1 variable.

The Default Behavior of cut

When you use the cut function without specifying any arguments, it defaults to several key settings. In our example, let’s see what happens when we call cut(data1) without any additional parameters.

data1cut <- cut(data1)

The output shows that the numbers have been divided into bins: 0-1, 1-2, 2-3, and so on.

However, notice how the upper bounds of these bins are not included in the classification. For example, the number 8 falls into the “7-10” bin because it’s greater than the lower bound of that range (7). If we want to include the upper boundaries in our classification, we need to specify this when calling cut.

Specifying Breaks

One of the key arguments of the cut function is breaks. This parameter allows us to define custom intervals for our data. In our example, let’s try specifying some breaks and see how it affects the output.

data1cut <- cut(data1, breaks = c(0,1,2,3,5,7,8,10), labels = FALSE)

Here, we’ve specified a vector of breaks that divides our data into seven distinct bins. Notice that we’ve used labels = FALSE, which means the function will return an integer vector representing the level codes for each bin.

Labeling Breaks

Another key argument is labels. By default, this parameter assigns labels to each bin based on the value of the breaks parameter. If you don’t want these default labels, you can set labels = FALSE, as shown in our previous example.

However, there’s an alternative: you can specify your own custom labels for the bins using the labels argument. For instance:

data1cut <- cut(data1, breaks = c(0,1,2,3,5,7,8,10), labels = c("Low", "Medium-Low", "Medium-High", "High-Med", "Med-High", "High", "Very High", "Extremely High"))

This code assigns custom labels to each bin, which can be useful for making the output more readable.

Including the Lower Bound

We mentioned earlier that by default, cut is exclusive of the lower bound. If you want to include it instead, you need to specify the include.lowest argument as TRUE.

data1cut <- cut(data1, breaks = c(0,1,2,3,5,7,8,10), labels = FALSE, include.lowest = TRUE)

With this adjustment, our classification becomes inclusive of the lower boundaries.

Summary Functions

Now that we understand how to use cut, let’s look at some summary functions. The summary function can be used with both raw data and binned data.

# Raw data summary
summary(data1)

Output:

Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
1.00    3.25    5.50    5.50    7.75   10.00 

# Binned data summary
summary(data1cut)

Output:

Min. 1st Qu. Median    Mean 3rd Qu.    Max. 
1.00    3.25    4.50    4.40    5.75    7.00  

As you can see, the two outputs are different. This is because summary functions provide summaries of the raw data and binned data separately.

Customizing Bins

To get a specific number of bins, we need to specify the breaks manually. Let’s try creating four bins based on our previous example.

data2cut <- cut(data1, breaks = c(1, 3.25, 5.50, 7.75, 10), labels = TRUE, include.lowest = TRUE)

Here, we’ve specified a vector of breaks and set labels = TRUE to use the default labels.

However, if you want more control over the labels or the ordering of the bins, you can define custom labels using character vectors. Alternatively, you can reorder the breaks using a numeric vector.

Reordering Breaks

Let’s say we want our first bin to start at 0 and our last bin to end at infinity. We can specify these breaks manually.

data3cut <- cut(data1, breaks = c(-inf, 0, 1, 2, 5, 7, inf), labels = TRUE)

Output:

 [1] (0-1]    (0-1]    (-inf,0]   (0,1]      (1,2)     (2,5)     (5,7)    
[7] (7, inf] 
Levels: (-inf,0] (0,1]  (1,2) (2,5) (5,7) (7, inf]

Note how the labels have been adjusted according to our custom breaks.

Conclusion

In this article, we explored the inner workings of the cut function in R and how it can be used to divide data into specified intervals. We discussed some key arguments, such as breaks, labels, and include.lowest, which allow users to customize their binned data outputs.

Last modified on 2023-06-01