In the previous chapter, you learned about discrete statistics methods for getting information about the distribution of discrete and continuous variables. In a data science project, the next typical step is to check for the associations between pairs of variables.

When checking for the associations between pairs of variables, you have three possibilities:

• Both variables are discrete
• Both variables are continuous
• There is one discrete and one continuous variable

Besides dealing with two variables only, this section also introduces linear regression, one of the most important statistical methods, where you model a single response (or dependent) variable with a regression formula that includes one or more predictor (or independent) variables.

Altogether, you will learn about the following in this section:

• Chi-squared test of the independence of two discrete variables
• Phi coefficient, contingency coefficient, and Cramer's V coefficient, which measure the association of two discrete variables
• Covariance and correlations between two continuous variables
• T-Test and one-way ANOVA, which measure associations between one continuous and one discrete variable
• Simple and polynomial linear regression