We are not statisticians by nature, but storytellers. Why don’t we make better use of that insight in our effort to predict and understand complex problems?
British economist John Kay presented the following problem in a recent Financial Times column When Storytelling Leads to Unhappy Endings:
Linda is single, outspoken and deeply engaged with social issues. Which of the following is more likely? That Linda is a bank manager or that Linda is a bank manager who is an active feminist?
If you chose the second answer you are in good company. Most of us do. Sadly, we’re wrong. Kay explains:
Many people say that the second option is more likely. Yet, the standard response goes, this cannot be. The rules of probability tell us the probability that both A and B are true cannot exceed the probability that either A or B is true. It is less likely that someone is a female Jamaican Olympic gold medalist than that a person is female, or that a person is Jamaican, or that a person is a gold medalist. Yet even people trained in probability make a mistake with the Linda problem. Or is it a mistake? Little introspection is required to understand what is going on. Respondents do not interpret the question as one about probability. They think it is a question about believability.
Believability, as Kay explains further, is narrative’s emblem. In the face of the messy, multi-faceted and open-ended situations that confront us, we humans tend to produce “simplifying narratives” that help make sense of events in a way we find believable, based on our personal, cultural and historical predispositions. Continue reading