STAT 118: Notes H

working with categorical data using forcats

Author

Affiliation

Emily Malcolm-White

Middlebury College

The R package forcats is designed to make working with categorical variables easier and more efficient. It provides a set of functions that allow you to manipulate and analyze categorical data with ease. In this lesson, we’ll cover the basics of the forcats package and some of its most useful functions.

Categorical Variables

Let’s review what categorical data is. Categorical data is a type of data that consists of categories or labels.

Examples of categorical data include:

Colors (red, blue, green, etc.)
Types of vehicles (sedan, SUV, truck)
Educational degrees (high school, college, graduate school)

Categorical data can be further divided into two types: nominal and ordinal. Nominal data consists of categories that have no inherent order, while ordinal data consists of categories that have a natural order. For example, educational degrees are ordinal data because they can be ordered from least to most advanced.

`mpg` Data

We will play with different functions in the forcats packages using the mpg dataset from earlier in the semester.

library(forcats)
library(tidyverse)
data("mpg")

Recall our side-by-side boxplot:

mpg %>% 
  ggplot(aes(x=class, y=hwy)) +
  geom_boxplot()

Reordering Factor Levels

One of the most useful functions is fct_relevel(), which allows you to reorder the levels of a factor. This can be useful when you want to change the default ordering of the levels or when you want to group certain levels together.

Is class a factor?

mpg$class %>% 
  is.factor()

[1] FALSE

Let’s make it a factor!

mpg$class <- mpg$class %>% 
  as.factor()

mpg$class %>% 
  is.factor()

[1] TRUE

Let’s check the levels and their current ordering!

mpg$class %>% 
  levels()

[1] "2seater"    "compact"    "midsize"    "minivan"    "pickup"    
[6] "subcompact" "suv"

To reorder the levels:

mpg$class <- mpg$class  %>% 
  fct_relevel("compact","subcompact","midsize","2seater","minivan","suv","pickup")

mpg$class %>% 
  levels()

[1] "compact"    "subcompact" "midsize"    "2seater"    "minivan"   
[6] "suv"        "pickup"

Let’s recreate our side-by-side boxplot now:

mpg %>% 
  ggplot(aes(x=class, y=hwy)) +
  geom_boxplot()

Rather than reordering them manually by typing the order, you could also re-level by some numeric criteria. For example:

mpg$class <- mpg$class %>% 
  fct_reorder(mpg$cty, median)

mpg$class %>% 
  levels()

[1] "suv"        "pickup"     "2seater"    "minivan"    "midsize"   
[6] "subcompact" "compact"

Renaming Factor levels

Sometimes you might not like the way the levels are named.

mpg$class <- mpg$class  %>% 
  fct_recode("two-seater" = "2seater")

## NEW NAME = OLD NAME

mpg$class %>% 
  levels()

[1] "suv"        "pickup"     "two-seater" "minivan"    "midsize"   
[6] "subcompact" "compact"

#Check out the change in the mpg dataset

Factor Collapsing

Let’s say we wanted to create only two categories – cars and larger vehicles.

mpg$class_two <- mpg$class %>% 
  fct_collapse(cars = c("compact", "subcompact", "midsize", "two-seater"),
               big = c("pickup", "suv", "minivan"))

mpg$class_two %>% 
  levels()

[1] "big"  "cars"

Lumping into an other category

fct_lump_min(): lumps levels that appear fewer than min times.
fct_lump_prop(): lumps levels that appear in fewer than (or equal to) prop * n times.
fct_lump_n() lumps all levels except for the n most frequent (or least frequent if n < 0)

table(mpg$manufacturer)


      audi  chevrolet      dodge       ford      honda    hyundai       jeep 
        18         19         37         25          9         14          8 
land rover    lincoln    mercury     nissan    pontiac     subaru     toyota 
         4          3          4         13          5         14         34 
volkswagen 
        27

Let’s say we wanted only the manufacturers with at least 15 cars produced. Everything else we want to just be other:

mpg$manufacturer <- mpg$manufacturer %>%   fct_lump_min(15)

mpg$manufacturer %>% 
  levels()

[1] "audi"       "chevrolet"  "dodge"      "ford"       "toyota"    
[6] "volkswagen" "Other"