No Credible Proof that Twitter Can Predict Elections, Researcher Finds
BY Nick Judd | Wednesday, May 2 2012
Computer science researcher Daniel Gayo-Avello writes in recently published research that many previous studies of Twitter's predictive power for elections have been "greatly exaggerated."
Technology Review points us to Avello's recent work, a survey of studies on Twitter statistics' correlation with election data. The researcher, at the University of Oviedo in Spain, writes:
... most of the authors seem to be more interested in claiming positive results than in providing sound and reproducible methods. It is also especially worrisome that many recent papers seem to only acknowledge those studies supporting the idea of Twitter predicting elections, instead of conducting a balanced literature review showing both sides of the matter. After reading many of such papers I have decided to write such a survey myself. Hence, in this paper, every study relevant to the matter of electoral prediction using social media is commented. From this review it can be concluded that the predictive power of Twitter regarding elections has been greatly exaggerated, and that hard research problems still lie ahead.
We vented about this in January, when Micah Sifry pointed out that the rising and falling of mention and follower counts are indicators of interest, not popularity.