Being wise after the event: an investigation of hindsight bias

Two experiments are reported which invesigate hindsight bias (the tendency to overestimate the probability of an event when one knows it has occurred and is asked to ignore the fact). Experiment I focuses upon the influence of commitments and desires upon subjective probability assessments in hindsight and foresight. Using the British General Election of May 1979 and Conservative and Labour Party members as subjects, only weak support for hindsight bias was found. However, party affiliation did exert a strong influence over likelihood assessments; outcomes favourable to the preferred party were perceived as more likely, in both foresight and hindsight, than outcomes favourable to the other party, and vice versa. Experiment 11 required subjects to make estimates in foresight and hindsight concerning the numbers of percentages of women in various roles in society. Strong evidence for hindsight bias was found. The paper concludes by considering the judgemental strategies subjects might have used in the two experiments. This is done by drawing on the heuristics of thinking proposed by Tversky and Kahneman.