For my sins, I write about health policy and try to translate research findings into prose readable by non-specialists. Put another way, I want to persuade you using data. But does statistical data persuade or motivate anyone?
I don't know of a study on this problem, but there is excellent and depressing work on how to persuade people to donate by Deborah Small, George Loewenstein, and Paul Slovic (for non-psychologists, this is a Bosch/Wade/James caliber team). The problem they studied is that people donate far more to help individual identifiable victims (the baby who has fallen down a well) than they will to help large group of unidentifiable 'statistical lives.'
OK, but surely adding statistical information to an emotional appeal would be the strongest way to motivate people to donate, right? Emotional narrative + Statistical evidence ought to be persuasive than either appeal alone?
What Small et al. found was perverse: adding statistical information to a narrative about an individual victim doesn't increase donation. It's actually worse than that:
if people are presented with a personal case of an identifiable victim along with statistical data about similar victims caught up in a larger pattern of illness, hunger or neglect, overall donations actually decline.
It's important not to overgeneralize from one line of research. Just because Small et al. did not find a way to present statistical information persuasively does not mean that no such way exists. Otherwise, it is as if the statistical information and deliberation about it poisoned the participants' altruistic emotive responses. Or perhaps that rational humans truly are egotists.