Module 05 of 1260 min readIntermediate

Multiple regression and partial effects

The Frisch-Waugh-Lovell theorem, controlling vs collider bias, and why 'controlling for X' is rarely as clean as it sounds.

42%

Module in draft

Long-form notes for this module are still being written.

In the meantime, head back to the course overview to see what is published, or jump to the next module if you have already covered this material.

← Previous

Module 04

Standard errors, p-values, and confidence

Next module →

Module 06

Endogeneity — the central problem

LeadAfrikPublic Economics Hub