Abstract

This paper investigates media attitudes towards migration in the aftermath of terrorist attacks and examines their impact on voting outcomes. Using natural language processing (NLP) methods, we analyze French news articles on immigration before and after three major terrorist events in 2015 and 2016. Our findings show that the salience of immigration in the French media increased for over seven weeks following the attacks. Notably, in the first three weeks, when immigration was linked to Muslims or (in)security-related issues, salience increased fivefold. Additionally, the tone of reporting became significantly more negative, though this effect was shorter-lived, lasting up to two or three weeks. In terms of electoral impact, we find that municipalities with higher exposure to negative media coverage saw an average increase of about 3% in votes for the main anti-migrant party during the Élections Régionales, held three weeks after the Bataclan attack. In some localities, however, the media’s influence was much greater, with contributions reaching as high as 20%.

Co-authors : Daniel Mirza , Matthieu Picault, Mamadou Sacko, and Ahmed Tritah

Practical information