Chapter 4. Correlation and Regression
Have you heard that ice cream consumption is linked to shark attacks? Apparently Jaws has a lethal appetite for mint chocolate chip. Figure 4-1 visualizes this proposed relationship.
Figure 4-1. The proposed relationship between ice cream consumption and shark attacks
“Not so,” you may retort. “This does not necessarily mean that shark attacks are triggered by ice cream consumption.”
“It could be,” you reason, “that as the outside temperature increases, more ice cream is consumed. People also spend more time near the ocean when the weather is warm, and that coincidence leads to more shark attacks.”
“Correlation Does Not Imply Causation”
You’ve likely heard repeatedly that “correlation does not imply causation.”
In Chapter 3, you learned that causation is a fraught expression in statistics. We really only reject the null hypothesis because we simply don’t have all the data to claim causality for sure. That semantic difference aside, does correlation have anything to do with causation? The standard expression somewhat oversimplifies their relationship; you’ll see why in this chapter using the tools of inferential statistics you picked up earlier.
This will be our final chapter conducted primarily in Excel. After that, you will have sufficiently grasped the framework of analytics to be ready to be move into R and Python.
Introducing Correlation ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access