We Shall Not Be Moved: The Surprising Non-Impact of Immigrant Violence

• In Event: Poster Session: Conflict Processes II

Abstract

How do close encounters with terrorism shape political sentiments? Contrary to the predictions of the existing literature, I find that attitudes moved little if at all for survivors of a 2016 terrorist attack in the United States. My data comes from a longitudinal panel of college students at multiple universities, a subset of whom personally experienced a terrorist attack on campus by a Muslim immigrant three weeks after Donald Trump’s election. Using a difference-in-differences estimator, I show that while there may be a marginal rightward shift on immigration by self-identified Republicans, the corresponding confidence intervals suggest that it is almost certainly small and is further attenuated when “Republican” is replaced with “Conservative.” Self-identified Democrats and liberals exhibit no such effects, a null result which remains precise under multiple model specifications and an alternative Bayesian approach. I also find party and ideological identification themselves to be stable, and register no differences in support for authoritarian or anti-minority policies nor political engagement, despite theories from the terrorism and crime victimization literature predicting such shifts. In addition to survey questions about stress and anxiety, I verify the incident’s emotional impact through interviews with the affected students. I propose a theoretical framework to explain when attitudes are likely to be impervious to such strong treatments and conclude by reflecting on how future terrorist attacks and high-profile violence by immigrants or minorities may influence—or fail to budge—a highly-polarized American electorate in the 2020 election and beyond.

Author

• Matthew D. Simonson, University of Pennsylvania