Study Significance
The paper proposes "agenda seeding," showing that marginalized activists can capture media and elite attention through disruption, thereby influencing public opinion and voting even when facing hostile majorities.
Measuring Protests and Political Outcomes
The analysis evaluates black-led protests from 1960–1972 and links local protest events to subsequent media coverage, congressional speech, public opinion, and changes in county-level presidential vote share.
The study contrasts nonviolent actions (for example sit-ins and marches) with protests that involved protester-initiated violence, and compares nearby counties’ political outcomes to assess how different tactics correlate with shifts in framing and voting.
Analysis and Estimation
A New Explanation for Influence Using rainfall as an instrumental variable via regression discontinuity design, this study quantifies that violent protests in 1968 likely shifted white voters' preferences toward Republicans by 1.5-7.9%. These findings suggest activists can indeed move agendas and discourse through strategic disruption.
Key Findings on Nonviolent versus Violent Protests
Nonviolent activism, especially when met with state or vigilante repression, increased media attention and framing favorable to civil-rights demands and was associated with a 1.6–2.5% increase in Democratic presidential vote share in proximate counties.
Protester-initiated violence also shifted news agendas and elite discourse, but toward themes of social control; in 1968 the author reports violent protests likely caused a 1.5–7.9% shift among whites toward Republicans and may have tipped the election.
