BookYol

Concept

Emotion Before Statistics

Emotion before statistics is the finding that one named person moves people more than a number describing thousands, and that adding the number to the person reduces the effect. The honest use is the true story first, then the figure that shows it is not alone.

People asked to donate after reading statistics about hunger gave about half what people shown one named child gave; people shown the child and the statistics together gave less than those shown the child alone. The researchers' explanation, which the Heaths report, is that numbers put people into an analytical frame of mind, and analytical people feel less. Kahneman's work adds that the fast, intuitive mind responds to the vivid individual and largely neglects base rates. The Heaths draw the craft from this: to make people care, make them feel something for one person before you show them the scale. This is also where honesty matters most, because the single vivid case is the manipulator's favourite tool, the anecdote standing in for a trend that does not exist. The honest persuader therefore carries both: the true story that lets the listener feel it, followed by the number that proves the story is representative, in that order, and only if the number actually supports the story. The defence when a story is used on you is the base-rate question: this is one case; how many are there, and how many went the other way?

A real-life example

A parents' association in Trabzon raising money for a school library stops opening its appeal with the number of books the school lacks. It opens instead with one pupil, named with her family's permission, who reads the same three books in turn because they are all her class has. Then it gives the figure: two hundred pupils in the same position. The appeal reaches its target within a month.

How to use it

  1. 1Open with one true person, told concretely and with permission, before any number.
  2. 2Follow with the figure that proves the case is representative; if the figure does not support the story, drop the story.
  3. 3When moved by a single case, ask how many there are and how many went the other way.