Population Sentiments and Violence: A Study of Emotional Temperature and National Homicide Rates, 2006–2019

The growing literature on the structural and cultural covariates of homicide rates contain few studies of the role of citizens' prevailing sentiments. Yet population sentiments and perceptions of daily life tell us something about societies beyond official measures of economic and demographic structure, and studies show perceptions of things like state legitimacy, civic engagement, and consumer sentiment are associated with national crime rates. Research also suggests collective emotions are associated with population-level outcomes like voting decisions, civil unrest, and prevalence of depression and anxiety. We studied the association between emotional temperature—a snapshot of citizens' positive and negative experiences—and national homicide rates. Our unit of analysis was the nation-year, 2006–2019. We obtained homicide data from the World Health Organization's Mortality Database. We gauged emotional temperature using the Gallup Global Emotions Report. The survey measures positive experiences (PEI) with questions about rest, respectful treatment, smiling or laughing, learning or doing something interesting, and enjoyment. It measures negative experiences (NEI) with questions about physical pain, worry, sadness, stress, and anger. We estimated separate associations for positive and negative experiences with pooled cross-sectional models, controlling for typical structural covariates and nation fixed effects. Results revealed no association between NEI and total, male, or female homicide victimization rates, nor between PEI and female victimization rates. However, we found that nation-years in which PEI is higher are nation-years in which total and male homicide victimization rates are lower. Additional analyses also suggest (1) an interaction effect such that both male and female homicide victimization rates may be higher when both NEI and poverty are higher and (2) a significant U-shaped curvilinear effect for NEI. Overall, our findings underline that citizens' shared experiences, and the population sentiments they produce, deserve greater attention in violence research.