Patrick Rafail, Tulane University
Social movements scholars have long analyzed media accounts of national protest activity drawn from a single, agenda-setting media outlet, especially the New York Times. Research has indicated that using a single national media source produces a highly biased depiction of mobilization, yet no major data collection efforts have attempted to both minimize the geographic biases of national media in combination with a large cross-section of national protest activity. This project addresses both of these issues: I use a unique research design to sample 11,858 protest events in 20 US cities based on 20 local newspapers between 1996 and 2006. These data were collecting using keyword sampling strategy that reduced more than 1 million candidate articles to a set of 10,953 articles that were manually coded. While labor intensive, this research design substantially reduces the geographic biases present in existing databases and generally increases the quality of information about the events themselves. The results point to significant, place-based variation in the initiating groups who stage events, the tactics that are used by protestors, police responses, and the scope of the claims made by activists. Based on these findings, I argue that protest activity is heavily informed by local structural, demographic, and socio-economic conditions. My approach provides a viable pathway for understanding where the distribution of who protests where, an issue that has received little systematic attention in recent scholarship on contentious political mobilization.
No extended abstract or paper available
Presented in Session 234. Empirical and Theoretical Advances in Protest Event Analysis