Speaker: Prof Nina Bobkova
Affiliation: Rice University
Zoom: https://uqz.zoom.us/j/82603079317
Abstract: From pharmaceutical marketing to political campaigns, communicators routinely choose between presenting aggregate data and highlighting an individual story. This paper asks when anecdotes versus statistics are more effective in persuading a heterogeneous audience. We model a policy advocate who seeks to persuade initially skeptical citizens to adopt a policy by presenting either local evidence—an anecdote about a single citizen’s outcome,—or global evidence—a statistic about the average policy impact. Citizens’ outcomes are correlated, with correlation declining sharply in demographic distance. Two forces shape the tradeoff: an umbrella effect, whereby the statistic has broader and more uniform reach, and a wild card effect, whereby the anecdote has higher dispersion and can generate extremely favorable realizations. We characterize when each type of evidence is optimal. Anecdotes are optimal when the required consensus is narrow or the audience is relatively homogeneous, whereas statistics are optimal when adoption requires broad consensus or the audience is more diverse.
About Economic Theory Seminar Series
A seminar series designed specifically for economic theory researchers to network and collaborate.