“Network Change Theory: Integrating Valente’s and Watts’ Models of Diffusion of Innovations”

Dr. David Krackhardt

Carnegie Mellon University

Abstract

Valente, following Rogers, has shown that identifying central actors is key to implementing or diffusing an innovation. Watts has argued that this strategy is ineffective, that instead one needs to access “susceptibles” in the network
and diffuse through large “Big Seed” strategies. I suggest a model that integrates both of these perspectives. This model makes the following claims: 1) Two different strategies operate on the micro and macro levels of diffusion. 2) On the micro level
(groups with less than Dunbar’s number of actors), it is possible to determine actors’ attributes on both dimensions of centrality (powerful) and susceptibility (supportive of the innovation). 3) An optimal strategy at the micro level is to start with those
high on both centrality and susceptibility (the innovators), then target those who are high on susceptibility but lower on centrality, then target those lower on susceptibility and centrality, and finally target those who are high on centrality but lower on
susceptibility (the resisters). 4) At the macro level, I use Viscosity Theory and Community Detection algorithms to find “islands” of actors within the larger system. 5) The strategy suggested by Viscosity Theory is to apply the micro strategy to one “island”
at a time, starting with an island that is peripheral (not central), that has relatively less interaction (fewer connections) with the rest of the network. Finally, Viscosity Theory also suggests macro structural conditions under which diffusion can happen
easily and conditions under which diffusion would be very expensive if not impossible.

More information: https://informatics.institute.ufl.edu/event/network-change-theory-integrating-valentes-and-watts-models-of-diffusion-of-innovations/

September 15th, 2016

11:30-1:30pm

Lunch will be provided!

UF Informatics Institute

432 Newell Drive, Rm E251, CSE Building RM E252

Gainesville. FL 32611