RationalWiki's 2018 Fundraiser

There is no RationalWiki without you. We are a small non-profit with no staff — we are hundreds of volunteers who document pseudoscience and crankery around the world every day. We will never allow ads because we must remain independent. We cannot rely on big donors with corresponding big agendas. We are not the largest website around, but we believe we play an important role in defending truth and objectivity.

If everyone seeing this today donates $5, we will meet our goal for 2018.

Fighting pseudoscience isn't free.
We are 100% user-supported! Help and donate $5, $20 or whatever you can today with PayPal Logo.png!

Donations so far: $3630Goal: $5000

Regression fallacy

From RationalWiki
Jump to: navigation, search
Part of the series on
Logic and rhetoric
Icon logic.svg
Key articles
General logic
Bad logic

A regression fallacy is a logical fallacy that occurs when an extreme value of some randomly varying event (something exceptional) is accepted as the normal value, and so when the value regresses to the mean, this change is believed to have been caused by some other event.

The fallacy is a causation fallacy and an informal fallacy.

Example[edit]

"Approximately 3,000 people were killed in the U.S. due to terrorism in 2001. There were twelve such deaths in 2002. This shows that techniques to prevent terrorism were massively improved after 2001."

This is fallacious. Historically, deaths due to terrorism in the U.S. are rare events, and therefore, after a major spike in the statistics, it is to be expected that there is a regression toward the mean: that the next year's figures would be more like the historical norm.

Explanation[edit]

In general, moving from a mere statistical correlation (or negative correlation) to proof of cause and effect is tricky. In the above example, anti-terrorist measures may indeed have played a role in preventing major attacks in 2002, but the statistics alone are only very weak evidence for this.

External links[edit]