Can social media predict voter intention in elections? The case of the 2012 Dominican Republic presidential election

Polls are the fundamental procedures used to measure voting trends in elections and a useful mechanism to adjust the marketing tactics of campaign commands. Polling not only demands significant resources, but also incorporates several intrinsic barriers that may hinder its applicability. This paper examines an innovative methodology to explore the use of social media as an alternative to enhance the costeffectiveness of polling while increasing significantly prediction power to overcome the constraints of traditional polls. The proposed methodology is based on the analysis of content gathered from public Web 2.0 technologies, and in this particular case, applied on the scenario of the 2012 Dominican Republic presidential election. Forecasting voting trends obtained from this methodology, and comparing such results with headline final election results, opens a promising road, not only to consolidate knowledge in this field, but also, as an new alternative for practitioners that might remarkably broaden electoral marketing strategies.