Yesterday we examined single-variable distributions. Today we explore relationships between variables. Do two variables move together? If so, how strongly and in which direction? And the riskiest question—does moving together imply one causes the other?
MLS-C01 Domain 2 treats correlation analysis as EDA's core tool. Correlation guides feature selection, multicollinearity diagnosis, and "should I include this feature?" decisions. Yet confusing correlation with causation is data science's costliest error.
A correlation coefficient quantifies how two variables change together