An emerging literature suggests that the success of social movements depends, partly, on their ability to garner support from third‐party groups. One factor that appears to predict support is social movements’ use of nonviolent (compared to violent) strategies to achieve their goals. However, this literature is not definitive. Herein, we report the results of a meta‐analysis of research that has assessed the effect of the use of nonviolence on third‐party support (k = 16, N = 4598). A small‐to‐moderate positive effect was observed, d = 0.25. Additionally, research that used a control or baseline comparison group suggested that using nonviolent strategies marginally (p = .090) increased people's willingness to help the movement (d = 0.17) while adopting violent strategies did not increase or decrease people's willingness to help the movement (d = −0.03). Publication bias was evidenced by bigger effect sizes of published (vs. unpublished) studies. Target (i.e., state vs. social issues) and location of the protest (i.e., domestic vs. foreign) were not significant moderators, whereas the context (i.e., real vs. hypothetical scenarios) was, although marginally. Results suggest that it behooves social movements to adopt nonviolent strategies if third‐party support is desired.