Correlation

In this section, we will explore the concept of correlation in more detail. As we saw in the previous section, correlation tells us how robust the relationship between two variables is. But, how do we know whether the correlation is statistically significant? If we want to make a prediction based on a linear regression model, we want to be sure that our relationship is meaningfully robust and that we are not just making a guess.

Typically, when we test the statistically significance of a correlation our null hypothesis is that two variables are not correlated with one another (\(r\) = 0) and we are testing whether or not to reject this null hypothesis. We will also consider the case where the null hypothesis is non-zero.

Let’s get started by reviewing what we know about correlation.