Lesson 3 of 4Module 1 · 0% of course
‹ PrevNext ›
Explore › Science › How Science Works › Method
How Science Works · Method

Experiments vs Correlation

Correlation doesn't prove causation because of confounders, reverse causation and chance. Randomized controlled experiments break those links. That's why they sit at the top of the evidence ladder.

35 min · 6 steps · videoJump to quiz ↓
Esther Duflo, Randomized Controlled Trials and Policy Making in Developing Countries MIT Institute for Data, Systems, and Society · 54 min
Open in YouTube
Step 1 of 6 · Overview35 min read ·
Key idea
Prereqs: The Scientific Method; Measurement & error. Next: Peer review & the replication crisis. Modeled on Khan Academy 'Statistical studies' (AP Stats) and Crash Course Statistics #8–9.
Next: 1. Why correlation misleads

Watch next

The Royal Institution · 63 min · works as audio
HISTORY · 45 min · works as audio
The People Profiles · 56 min · works as audio
The Royal Institution · 60 min · works as audio
YaleCourses · 43 min · works as audio
The Royal Institution · 86 min · works as audio
MIT OpenCourseWare · 54 min · works as audio
Biographics · 28 min · works as audio
Dwarkesh Patel · 55 min · works as audio
Spark · 89 min · works as audio
YaleCourses · 68 min · works as audio
MIT Institute for Data, Systems, and Society · 54 min · works as audio
MIT OpenCourseWare · 49 min · works as audio
The Cārvāka Podcast · 53 min · works as audio
The People Profiles · 64 min · works as audio
StarTalk · 42 min · works as audio
‹ PrevNext ›