Measuring Rural Economic Hardship and Its Association with Intimate Partner Homicide: Implications for Future Research

Criminologists have long examined the relationship between community-level factors and crime, including homicide. However, these studies have rarely disaggregated by homicide type or by place, or have focused solely on urban communities. Intimate partner homicide (IPH) has been shown to have different antecedents at the individual-level than general homicide, and there is some evidence that community-level characteristics which increase IPH risk may differ as well, although this is a nascent line of inquiry. More research that is attentive to community characteristics’ effects on IPH rates and attends to place is needed. This presentation will draw from two studies to discuss the measurement of community-level factors, including rurality and economic hardship, to discuss future research directions. The first study used a multilevel model to test the extent to which individual-level (e.g., victim sex), county-level (e.g., unemployment rate), and state-level (e.g., firearm prevalence) characteristics associated with IPH in a national sample of rural counties between 2009 and 2016 (“national study”). The second study study utilized spatial and non-spatial analytic techniques to examine how the clustering of community characteristics predicted intimate partner homicide (IPH) rates in Massachusetts' towns between 2005 and 2014 (“Massachusetts study”). In both studies, a higher percentage of households in the community receiving public assistance was associated with IPH (national study, ß= 0.148) and IPH rate (Massachusetts study, IRR= 1.22). In the national study, higher unemployment rates were associated with non-IPH, suggesting that in rural areas, other measures of community economic hardship may be more related to IPH (e.g., underemployment) than typical measures used in past studies (e.g., unemployment rate, percent of the population below the poverty line). In the Massachusetts study, a 10% increase in the proportion of a town’s population that lived in rural areas (<500 people per square mile) was associated with a 13% increase in the IPH rate. This suggests that future studies of community-level characteristics’ effects on IPH would benefit from including measures of rurality in their models or disaggregating by place. Implications for measurement of community-level characteristics in future studies and avenues of future inquiry will be discussed.