Working Paper
Distinguishing among Mechanisms of Social Contagion in New Product Adoption: Framework and Illustration
Raghuram Iyengar, Christophe Van den Bulte, and Jeonghye Choi, 2011 [11-119]
View/Order Item >>
Comments from members >>
Marketers have become very keen on leveraging social influence among customers. One form of influence that has received much attention is “social contagion”—that people’s decision to adopt a new product or technology is affected by the extent to which their peers are already using it.
Research on how new products gain market acceptance is moving from investigating
whether to
why contagion is at work. To answer that question, one must be able to distinguish among multiple contagion mechanisms in real market settings rather than just within consumer laboratories. In this report, authors Iyengar, Van den Bulte, and Choi propose a framework to do so.

To begin, they suggest that different contagion mechanisms operate through different stimuli, through different sources or conduits, or with different strength across products, market conditions, and people. They apply this contingency-based framework to distinguish between learning about the product’s benefits and risks (“social learning”) and adherence to norms of proper behavior (“normative influence”).
Their research setting is the adoption of a new prescription drug used by physicians to treat a chronic and potentially life-threatening medical condition. Leveraging extant theory and research findings, the authors propose that social learning and normative influence operate through different sources or conduits (discussion/referral ties versus immediate colleagues) and through different stimuli (share versus number of prescriptions), and that social learning is moderated by whether a physician feels he or she can learn from others.
There are four key findings in the context of their study:
- Both immediate colleagues and discussion/referral partners (a majority of which were outside the physicians’ workplace) were sources of social contagion.
- Contagion from immediate colleagues was based on the extent to which they were committed to the new drug, as reflected by its share in their prescriptions in the category, and was not affected by their experience with the new drug as reflected by prescription volume.
- In contrast, contagion from discussion and referral partners was based on their experience with the new drug (prescription volume) rather than on their commitment to the new drug (share in category).
- The extent to which a physician feels he or she cannot learn much from others (self-reported opinion leadership) depresses contagion from discussion and referral partners, but does not affect contagion from immediate colleagues.
Those findings suggest that social influence operated through both social learning and normative influence. Understanding the broad mechanisms driving contagion has great relevance to marketing practice. The nature of the mechanism at work not only affects aggregate-level diffusion curves, but should also inform marketing decisions about which customers to target as seeding points and which ties to activate with a particular message or appeal.
Raghuram Iyengar is Assistant Professor of Marketing and Christophe Van den Bulte is Associate Professor of Marketing, both at the Wharton School of the University of Pennsylvania. Jeonghye Choi is Assistant Professor of Marketing, Yonsei School of Business, Yonsei University, Korea.
Acknowledgments
We thank Thomas Valente as well as John Eichert and Bruce West at Rivermark and several managers and analysts at the participating company for their efforts in making this research possible. We benefited from a discussion with David Krackhardt and from comments by Jonah Berger, John Eichert, Blake McShane, participants at the 2009 INFORMS Marketing Science Conference, the 2009 Collaborative and Multidisciplinary Research Conference organized by the Center for Customer Insights at Yale University, the 2010 Marketing Research Camp at The Pennsylvania State University, and seminar participants at Emory University, the University of Maryland, Massachusetts Institute of Technology, New York University, and Temple University.
Included in the following collections:
MSI Reports 2011
Comments from members
Please
login to view and/or submit comments
View/Order Item