Influence and Attention on Twitter - Duncan Watts
Duncan James Watts, Principal Researcher at Microsoft Research, New York City
Relevant papers:
- "Who Says What to Whom on Twitter"
- "Everyone’s an Influencer: Quantifying Influence on Twitter"
Abstract
Word-of-mouth diffusion of information is of great interest to planners, marketers and social network researchers alike. In this talk I discuss two recent studies of Twitter users. In the first study, I describe some results about the distribution of attention on Twitter between categories of “elite” users, identified using Twitter lists, and “ordinary” users. These results also shed some interesting new light on the classical theory of the “two-step flow,” one of the dominant theories of 20th Century communications research. The second study considers 90 million diffusion events that took place on the Twitter follower graph over a two-month interval in 2009. Although the largest cascades tend to be generated by users who have been influential in the past, and from urls that were rated more interesting and/or elicited more positive feelings, predictions of which user or URL will generate large cascades are unreliable. The main upshot of this result is that word-of-mouth diffusion can only be harnessed reliably by targeting large numbers of potential influencers, thereby capturing average effects. In addition, I also consider a family of hypothetical marketing strategies, and argue that under a wide range of plausible assumptions the most cost-effective performance can be realized using relatively 'ordinary influencers'---individuals who exert average or even less-than-average influence.