6 Categorical data

 

This chapter covers

  • Determining the best approach to handle categorical data
  • How to avoid common mistakes when working with categorical data
  • Analyzing categorical data with the right methods to investigate patterns and associations

Sometimes, you may encounter data that is not numeric, and your typical analysis methods won’t apply. Instead, this data represents groups or categories with values limited to a set number of options. Customer locations, departments, and demographics are typical sources of such discrete values. We call this categorical data, and it is common in most datasets.

Methods suitable for numeric or continuous data are not appropriate for categorical data. Knowing the correct way to handle categorical data will broaden your toolbox and ensure you use the right tool for the job when your data is mostly categorical. In this chapter, we will review common tools for handling categorical data before diving into the project.

6.1 Working with categorical data

6.1.1 Methods for handling categorical data

6.1.2 Working with survey data

6.2 Project 5: Analyzing a survey to understand developer attitudes toward AI tools

6.2.1 Problem statement

6.2.2 Data dictionary

6.2.3 Desired outcomes

6.2.4 Required tools

6.3 Applying the results-driven method to analyzing the developer survey

6.4 An example solution: How do developers use AI?

6.4.1 Exploring categorical data

6.4.2 Analyzing categorical survey data

6.4.3 Project progress so far

Summary