Valentina Assenova, Management, The Wharton School, and Nidhi Sridhar, Computer and Information Science, University of Pennsylvania
Abstract: The networks literature examining the diffusion of complex innovations by social contagion has focused on the benefits of “multiplex” (or “wide”) ties in these processes. Multiplex ties span different types of networks to create inter-connectivity across subgraphs within a community. Prior research has postulated that multiplex ties promote diffusion by enabling greater social validation for a complex innovations, but existing literature has not distinguished between different types of multiplex ties. In this article, we propose a typology of multiplex ties: ties that span structurally similar subgraphs (“homogeneous multiplex”) and ties that span structurally different subgraphs (“heterogeneous multiplex”). We demonstrate that network-based social contagion processes unfold differently across these types. We further demonstrate that the prior predictions that multiplexity promotes diffusion hold only when multiplex ties span structurally dissimilar networks. Drawing on a longitudinal dataset from the diffusion of microfinance in 43 Indian villages, we first characterize the network topologies into distinct types of multiplex ties, and then show that these types correspond to different subgraph properties and rates of diffusion. Our findings surface important limitations to the benefits of multiplexity for the diffusion of complex innovations, by showing that these benefits are contingent on network content.