The Internet is thick with social networks, but it is not clear which conditions favor the rapid spread of information or the adoption of behaviors offline. One theory maintains that an online network with more “weak ties” and overall connections can quickly and efficiently encourage the adoption of a given social behavior. A competing theory suggests that a network with more clusters of overlapping connections can better promote that behavior by delivering the same message to an individual multiple times.
A 2010 study from MIT published in Science, “The Spread of Behavior in an Online Social Network Experiment,” compared the effectiveness of network clusters versus a more diffused one comprised of both weak and strong ties in spreading a given behavior. Researchers created a closed online health network and assigned some 1,500 participants to one of two network types: a clustered-lattice network characterized by multiple links to other participants, or a homogenous network comprised of random associations and no overlapping links. Researchers then tracked who chose to sign up for an online health forum that was promoted only through messages from network contacts.
Key study findings include:
The author suggests that “interventions aimed at the spread of new health behaviors (for instance, improved diet, regular exercise, condom use, or needle exchange) may do better to target clustered residential networks rather than the casual contact networks across which disease may spread very quickly.”
A 2011 related study from Harvard University published in the Proceedings of the National Academy of Sciences, “Dynamic Social Networks Promote Cooperation in Experiments with Humans,” reinforced the MIT study’s findings — namely, that a clustered social network can be highly effective method for distributing information and promoting behavioral changes. However, the Harvard study cautions that cooperative behaviors decay over time and that a successful social network both introduces new members and expunges less productive connections.
Tags: Facebook, Twitter, technology