Comparing Social Media and Traditional Surveys Around the Boston Marathon Bombing

Sociological surveys have been a key instrument in understanding social phenomena, but do the introduction and popularity of social media threaten to usurp the survey's place? The significant amount of data one can capture from social media sites like Twitter make such sources appealing. Limited work has tried to triangulate these sources pragmatically for research. This paper documents experiences in comparing analyses and results from a panel survey, a survey embedded within an experiment, and social media data surrounding the 2013 Boston Marathon Bombing. Our experience suggests the sources are complementary: social media provides better insight into behavior more rapidly and cheaper than surveys, but surveys can provide higher quality, targeted, and more relevant data.