StatQuest with Josh Starmer on YouTube · Open on YouTube
Core skills: SQL depth, statistics, Python and dashboards · Advanced beginner · Lesson 21 of 99 · 27 min video, then about 4 hours of practice
Correlation and simple regression for analysts
After this you can fit and interpret a simple linear regression, read R-squared and coefficients, and explain why correlation does not establish causation.
Why it matters
Regression is how analysts quantify relationships ('each extra support ticket is associated with a 2 percent lower renewal rate'). Focus on interpreting coefficients in business units, reading R-squared without overvaluing it, and checking residuals. The trap is treating a regression coefficient as a causal effect; you will learn to reason about causation properly in the experimentation phase.
How to use this resource
StatQuest's full explanation of linear regression, R-squared and p-values for the fit.