6.1 Introduction

Automating bivariate analyses (both stats and charts) is similar to univariate statistics. However, you will need to always identify the label in each dataset to reduce the number of analyses that are generated and focus on the target. This book categorized bivariate analyses and charts based on the data type of the feature and the label. Therefore, a good bivariate function will calculate the appropriate statistical test (r, X², or F) for each of the four cases summarized in the table below.

In the prior chapter, you learned the principles of creating useful functions that are:

• Automated

• Dynamic

• Error-resistant

The prior chapter on univariate automation focused mostly on making the functions automated and dynamic. In this chapter, we will extend the range of dynamism and begin some basic error-resistance by building out a few functions together and including some data cleaning. The practice problems and assessment will encourage you to create more functions on your own.