Wells effect

Quick Reference

A reluctance to make judgements of legal liability on the basis of naked statistical evidence alone. In the original experiments, evidence from tyre tracks providing an 80 per cent probability that a particular bus company was responsible for running over a dog persuaded fewer than 20 per cent of experienced judges and people without legal training to rule against the bus company, although they were instructed to decide on the balance of probabilities. In contrast, evidence that the particular bus company was responsible for running over the dog, based on an analysis of tyre tracks that was said to be 80 per cent reliable, persuaded almost 70 per cent to rule against the company. Thus, evidence that is 80 per cent reliable is sufficient to persuade most people, but naked statistical evidence of an 80 per cent probability is not, although the mathematical probability is the same in both cases. See also base-rate fallacy, subjective probability, taxicab problem. [Named after the US psychologist Gary L(eroy) Wells (born 1950) who first drew attention to the phenomenon in 1992]

Subjects: Psychology.

Reference entries