The present study (N = 1074) examined the impact of a theory‐driven media intervention aimed at violence prevention and intergroup reconciliation in Burundi. We used a novel methodology utilizing audio‐based surveys to assess attitudes related to intergroup conflict and reconciliation among community members. We conducted a propensity score analysis to estimate the causal effects of the intervention by examining differences between listeners and non‐listeners of the radio dramas. The results indicated a positive effect of the intervention on several social psychological outcomes (tolerance, in‐group superiority, social distance, intergroup trust, responsibility attributions, trauma disclosure and competitive victimhood). However, listeners and non‐listeners did not differ in obedience toward leaders or historical perspective taking; and the results for active bystandership, one of the main foci of the intervention, were mixed. Furthermore, the results show that the impact of the intervention sometimes depends on listeners' personal experiences of victimization. Copyright © 2015 John Wiley & Sons, Ltd.