Tie Strength, Embeddedness & Social Influence: Evidence from a Large Scale Networked Experiment

30 Pages Posted: 9 Jan 2013 Last revised: 25 Jan 2015

See all articles by Sinan Aral

Sinan Aral

Massachusetts Institute of Technology (MIT) - Sloan School of Management

Dylan Walker

Chapman University

Date Written: January 8, 2013


Understanding peer influence in networks is critical to estimating product demand and diffusion, creating effective viral marketing, and designing ‘network interventions’ to promote positive social change. But several statistical challenges make it difficult to econometrically identify peer influence in networks. Though some recent studies use experiments to identify influence, they have not investigated the social or structural conditions under which influence is strongest. We investigate the two most prominent network characteristics that may moderate social influence between peers - tie strength and network embeddedness. By randomly manipulating messages sent by adopters of a Facebook application to their 1.3 Million peers, we were able to identify the moderating effect of tie strength and embeddedness on influence. We find that both embeddedness and tie strength increase influence. Individuals experience a 0.6% increase in influence over their peers for each friend they share in common with that peer. As the number of common friends can be quite large, this effect is also economically significant. Individuals exert 125% more influence on peers for each affiliation they share in common, 1355% more influence on peers with whom they attended the same college, and 622% more influence on peers that live in the same current town. However, the amount of physical interaction between friends, measured by co-appearance in photos, does not have an effect. This work presents some of the first large scale experimental evidence investigating the social and structural moderators of peer influence in networks. The results could enable more effective marketing strategies and public policy more broadly.

Suggested Citation

Aral, Sinan and Walker, Dylan, Tie Strength, Embeddedness & Social Influence: Evidence from a Large Scale Networked Experiment (January 8, 2013). Available at SSRN: https://ssrn.com/abstract=2197972 or http://dx.doi.org/10.2139/ssrn.2197972

Sinan Aral (Contact Author)

Massachusetts Institute of Technology (MIT) - Sloan School of Management ( email )

100 Main Street
Cambridge, MA 02142
United States

Dylan Walker

Chapman University ( email )

1 University Drive
Orange, CA 92866
United States

HOME PAGE: http://https://dylantwalker.com

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
PlumX Metrics