Did They Tell Their Friends? - Using Social Network Analysis to Detect Contagion Processes
Kühne Logistics University - THE KLU; Christian-Albrechts-University at Kiel, Institute for Innovation Research
Christian-Albrechts-Universitaet at Kiel
Social contagion processes such as word-of-mouth (WOM) are widely regarded as key success factors for innovation diffusion. Aspects of these processes have been thoroughly explored in empirical studies on the actor or dyad levels of analysis. While such studies offer valuable insight into the motivations and contents of WOM, they are not able to include social network structures in their analysis. Contagion processes, however, require an underlying social networks infrastructure to unfold their potential for innovation diffusion. Although marketing managers strongly believe in social contagion processes, and studies on both actor and dyad levels strongly suggest their existence, marketing scientists have been unable to find conclusive evidence of such effects in network-level studies, which are most appropriate for this purpose.
To address this research gap, we propose a new approach for empirical research in this field: the quantitative determination of communication activity, reach, speed, and epidemicity of observed diffusion processes by using social network analysis. We apply this approach empirically by analyzing anonymized customer data from an innovative telecommunications provider. The resulting social network of adopters, consisting of 55,065 customers and 7.8 million individual phone calls, is analyzed and compared to simulated random networks of similar dimensions. We find strong support for significant social contagion influences on adoption decisions and an epidemic pattern of innovation diffusion.
Number of Pages in PDF File: 39
Keywords: Diffusion, Social Networks, Telecommunications, Social Contagion, Word-of-Mouthworking papers series
Date posted: February 12, 2008
© 2015 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo3 in 1.266 seconds