In this chapter, you’ll learn all about factors. You’ll discover the difference between categorical and ordinal variables, how R represents them, and how to inspect them to find the number and names of the levels. Finally, you’ll find how forcats, a tidyverse package, can improve your plots by letting you quickly reorder variables by their frequency.

Introduction to qualitative variables

Recognizing factor variables

Qualitative variables in theory

Understanding your qualitative variables

Getting number of levels

Examining number of levels

Examining levels

Making better plots

Reordering a variable by its frequency

Ordering one variable by another

Introduction to Factor Variables

You’ll continue to dive into the forcats package, learning how to change the order and names of levels and even collapse them into one another.

Reordering factors

Changing the order of factor levels

Tricks of fct_relevel()

Renaming factor levels

Distinguishing between forcats functions

Renaming a few levels

When you have a typo

Collapsing factor levels

Manually collapsing levels

Lumping variables by proportion

Preserving the most common levels

Manipulating Factor Variables

Having gotten a good grasp of forcats, you’ll expand out to the rest of the tidyverse, learning and reviewing functions from dplyr, tidyr, and stringr. You’ll refine graphs with ggplot2 by changing axes to percentage scales, editing the layout of the text, and more.

Examining common themed variables

Grouping and reshaping similar columns

Summarizing data

Creating an initial plot

Tricks of ggplot2

Editing plot text

Reordering graphs

Changing and creating variables with case_when()

case_when() with single variable

case_when() from multiple columns

Creating Factor Variables

In this final chapter, you’ll take all that you’ve learned and apply it in a case study. You’ll learn more about working with strings and summarizing data, then replicate a publication quality 538 plot.

Case study introduction

Changing characters to factors

Tidying data

Data preparation and regex

Cleaning up strings

Dichotomizing variables

Recreating the plot

Fixing labels

Flipping things around

Finalizing the chart

End of course recap

Case Study on Flight Etiquette

538 Flying Etiquette survey

Kaggle multiple choice responses

As a data scientist, you will often find yourself working with non-numerical data, such as job titles, survey responses, or demographic information. R has a special way of representing them, called factors, and this course will help you master working with them using the tidyverse package forcats. We’ll also work with other tidyverse packages, including ggplot2, dplyr, stringr, and tidyr and use real world datasets, such as the fivethirtyeight flight dataset and Kaggle’s State of Data Science and ML Survey. Following this course, you’ll be able to identify and manipulate factor variables, quickly and efficiently visualize your data, and effectively communicate your results. Get ready to categorize!

Reshaping Data with tidyr

Get ready to categorize! In this course,  you will work with non-numerical data, such as job titles or survey responses, using the Tidyverse landscape.

Categorical Data in the Tidyverse

Tidyverse Grundlagen in R

Renaming a few levels

Categorical Data in the Tidyverse

Interaktive Übung