The Texas sharpshooter fallacy is a logical fallacy where a cluster of statistically non-significant data is taken from its context, and therefore thought to have a common cause.
The name comes from a story about a Texan who fires his gun randomly at the side of a barn, then paints a target centered on the largest cluster of hits.
The fallacy is closely related to the clustering illusion, which refers to the tendency in human cognition to interpret patterns in randomness where none actually exist.
Examples
- Attempts to find cryptograms in the works of William Shakespeare, which tended to report results only for those passages of Shakespeare for which the proposed decoding algorithm produced an intelligible result.
- The roulette ball has landed on odd numbers eight times in a row now... there must be something wrong with it. (The belief that after eight odd numbers in a row the ball is more likely to land on an even number is an example of gambler's fallacy.)
- More children in town A have leukemia than in town B. Therefore, there must be something wrong with town A. (If that fact is linked to another one, it may be an example of cum hoc ergo propter hoc or post hoc ergo propter hoc.)
Counterexamples
One should be cautious not to dismiss a set of events which may have a shared underlying physical cause as being due to "fallacy." For example, engineers were aware of a problem with burnt o-rings that form the inter-segment seals for the solid rocket boosters for the Space Shuttle prior to the Challenger accident. In particular it was known that the burn-through problem was worse with lower temperatures at launch. Postulating a physical mechanism, one might infer that the o-ring material shrinks or becomes brittle and thereby fails to seal at the moment of launch. With a postulated mechanism, the dataset of several burn-through incidents among twenty-four previous launches becomes highly significant, particularly when ambient temperature is taken as the controlling parameter .
See also
External links
Last updated: 05-07-2005 07:16:07
Last updated: 05-13-2005 07:56:04
