Presented at the CSED Seminar Series
Innovations often spread by the communication of information among potential adopters. In the marketing literature, the standard model of new product diffusion is generated by information contagion: agents adopt once they hear about the existence of the product from someone else. In social learning models, by contrast, an agent adopts only when the perceived advantage of the innovation — as revealed by the actions and experience of prior adopters — exceeds a threshold determined by the agent’s prior beliefs. We demonstrate that learning with heterogeneous priors generates adoption curves that have an analytically tractable, closed-form solution. Moreover there is a simple statistical test that discriminates between this type of process and a contagion model. Applied to Griliches’ classic results on the adoption of hybrid corn, this test shows that learning with heterogeneous priors does a considerably better job of explaining the data than does the contagion model.
"I think the power of #MeToo is how it reveals the overwhelming scope and breadth of these problems, and how they affect victims. It forced individuals to recognize that there are structural features to what’s happening, and thus that everyone has a role to play in preventing assault and harassment.”