Post Hoc Ergo Propter Hoc (False Cause)
Assuming that because event B followed event A, event A caused event B. Latin for “after this, therefore because of this.” Correlation does not imply causation.
Examples
- “I wore my lucky socks and my team won. My socks caused the win.”
- “Crime rates dropped after the new mayor took office. The mayor’s policies reduced crime.”
- “Autism diagnoses increased after vaccine schedule expanded. Vaccines cause autism.” (Debunked — timing coincides with diagnostic changes)
- “I took this supplement and my cold went away. The supplement cured me.” (Colds resolve on their own)
Why It’s Seductive
- Temporal sequence is easy to observe; causal mechanisms are hard to verify
- Human brains are pattern-seeking machines
- We underestimate coincidence and regression to the mean
Valid Causal Inference Requires
- Temporal precedence: Cause precedes effect
- Covariation: Cause and effect vary together
- No plausible alternative explanations (controlled for confounders)
- Mechanism: A plausible causal pathway
How to Counter
- “What else changed at the same time?”
- “Is there a control group or natural experiment?”
- “Could this be regression to the mean?”
- “What’s the proposed mechanism? Is it plausible?”
- “Would the effect have happened anyway?”