# 33. Statistical manipulation - False causality ## 33.5. Methodology/Refinements/Sub-species ### 33.5.2. Clustering illusion, aka the Texas sharpshooter fallacy This is a fallacy in which pieces of information that have no relationship to one another are highlighted because of their similarities, and that similarity is used to claim the existence of a pattern or correlation. The name comes from the joke about a Texan who fires some shots at the side of a barn, then paints a target centred on the biggest cluster of hits and claims to be a sharpshooter. #### 33.5.2.1. Manipulative uses In manipulation, a user claims that they take randomness into account when determining cause and effect, whereas in reality they do not. They may even deliberately choose a cluster to match their manipulative intent. The manipulation works because, in reality, we all tend to ignore random chance when results seem meaningful, or when we (emotionally) want a random effect to have a meaningful cause. This irrational human tendency is known and used by the manipulator to prove a manipulative case to the victim. #### 33.5.2.2. Psychological origins and statistical explanations In psychology, the effect is called "the clustering illusion". It describes the tendency in human cognition to interpret patterns in randomness where none actually exist. The clustering illusion is the intuition that random events which occur in clusters are not really random events. The illusion is due to selective thinking, based on counterintuitive and false assumptions regarding statistical odds (see "The Gambler's Fallacy"). This is the idea that because, when tossing a coin, there is a 50% chance of getting heads or tails, that after 100 tosses of the coin giving heads, there is somehow a higher chance of getting tails. There isn't. The clustering of a result like this in the short term is not at all unusual in nature, and this clustering phenomenon can be used to manipulatively imply or "demonstrate" some illicit conclusion to an unsuspecting victim. For example, in epidemiological studies of cancer, finding a statistically unusual number of cancers in a given neighbourhood - such as six or seven times greater than the average - is not that rare or unexpected. Much depends on where you draw the boundaries of the neighbourhood and the demographics of a neighbourhood; these can radically affect the incidence. However, clusters of cancers that are seven thousand times higher than expected, such as the incidence of mesothelioma in Karian, Turkey, are very rare and unexpected. The incidence of thyroid cancer in children near Chernobyl was one hundred times higher after the disaster. Such clusters are the result of a real environmental cancer risk. In epidemiology, Khaneman and Tversky called the clustering illusion the "belief in the Law of Small Numbers", because they identified the clustering illusion with the fallacy of assuming that the pattern of a large population will be replicated in all of its subsets. Logically, this fallacy is known as the "fallacy of division" because it assumes that the parts must be exactly like the whole.