The Texas sharpshooter fallacy is an informal fallacy committed when a false conclusion is reached by focusing on evidence which supports it, whilst disregarding that which does not. This fallacy is the philosophical or rhetorical application of the multiple comparisons problem (in statistics) and apophenia (in cognitive psychology). It is related to the clustering illusion, which is the tendency in human cognition to interpret patterns where none actually exist.

The name comes from a metaphor about a person from Texas who fires a gun at the side of a barn, then paints a shooting target centered on the tightest cluster of shots and claims to be a sharpshooter.

Structure

thumb|A set of 100 randomly generated points displayed on a scatter graph. Examining the points, it is easy to identify apparent patterns. In particular, rather than spreading out evenly, it is not uncommon for random data points to form clusters, giving the (false) impression of "hot spots" created by some underlying cause.

The Texas sharpshooter fallacy often arises when a conclusion is based on the analysis of a highly restricted subset of available data. Some factor other than the one attributed may give all the elements in that subset some common property (or pair of common properties, when arguing for correlation). If the person attempts to account for the likelihood of finding some subset in the large data with some common property by a factor other than its actual cause, the person is likely committing a Texas sharpshooter fallacy.

The fallacy is characterized by the failure to specify a hypothesis before gathering data, or to formulate one only after data has been collected and reviewed (HARKing). Thus, it typically does not apply if one had an ex ante, or prior, expectation of the particular relationship in question before examining the data. For example, before examining the information, one might have in mind a specific physical mechanism implying the particular relationship. One could then use the information to give support or cast doubt on the presence of that mechanism. Alternatively, if a second set of additional information can be generated using the same process as the original information, one can use the first (original) set of information to construct a hypothesis, and then test the hypothesis on the second (new) set of information. (See hypothesis testing.) However, after constructing a hypothesis on a set of data, one would be committing the Texas sharpshooter fallacy if they then tested that hypothesis on the same data (see hypotheses suggested by the data).

Examples

In 1993, Swedish researchers reported the results of study designed to investigate whether long-term proximity to high-voltage power lines was linked to negative health outcomes. Health surveys of people living within 300 metres of such power lines were undertaken over a 25-year period, following which the data collected was analysed in order to determine whether a statistically significant increase could be detected in the prevalence of over 800 medical conditions, when compared to people who lived further away. The study found that the incidence of childhood leukemia was four times higher in the former group, prompting calls for action by the Swedish government. The problem with the conclusion, however, was that the number of potential ailments, i.e., over 800, was so large that it created a high probability that at least one ailment would have a statistically significant correlation with living distance from power lines by chance alone, a situation known as the multiple comparisons problem. Subsequent studies failed to show any association between power lines and childhood leukemia.

The fallacy is often found in modern-day interpretations of the quatrains of Nostradamus. Nostradamus's quatrains are often liberally translated from their original Middle French versions, in which their historical context is often lost, and then applied to support the erroneous conclusion that Nostradamus predicted a given modern-day event after the event actually occurred.

See also

  • , also known as p-hacking
  • '
  • '

References

  • Fallacy files entry