Like ordinary crime, terrorism hot spots are predominately located in large, metropolitan areas. While some locales remain targets of terrorist attacks, to a large extent hot spots of terrorist attacks demonstrate a significant amount of variability over time. Moreover, we find significant variability in the ideologies motivating terrorist attacks across decades.
Terrorism and ordinary crime occur in many of the same areas. We find that while some traditional predictors of ordinary crime also predict terrorist attacks, many robust correlates of ordinary crime do not. These data were limited in some respects; much more work in this area is needed to fully understand the linkages between terrorism and ordinary crime.
The goal of this research was to fill a gap in the terrorism literature by documenting descriptive patterns of terrorist attacks in the United States over time and space. While terrorism has recently received much public attention, the patterns documented here show that U.S. terrorist attacks have been relatively infrequent in the last decade.
While it is encouraging to find that terrorist attacks are down from the highs experienced in the 1970s and have maintained a low level for some time, we also see a rise in the likelihood of fatalities among recent terrorist acts. Moreover, there is evidence that there have been large increases in the proportion of foiled to completed plots in the last decade (Dahl, 2011). This is important because the GTD does not include foiled plots where no specific action had yet been initiated.
Similar to ordinary crime, certain counties can be characterized as hot spots of terrorist attacks. Moreover, like ordinary crime, terrorism hot spots are dominated by large metropolitan areas such as Los Angeles County, CA, Miami-Dade County, FL, and Manhattan County, NY. The clustering of terrorist attacks in large urban areas was consistently documented across the entire 1970 to 2008 time period. Notably, although large, urban areas are a prime location for terrorism hot spots, the identification of hot spot locations differs substantially across decades. That is, while some locales seem to remain prime targets of terrorist attacks from the 1970s through today (i.e., Los Angeles, Manhattan), for the most part hot spots change across decades. San Francisco County, CA, was a prime target in the 1970s whereas Maricopa County, AZ, has recently become a hot spot for terrorism. This patterning is likely due to changes in ideological motivation over this same time period.
Whereas the 1970s were characterized by terrorist attacks motivated by extreme left-wing and ethno-national/separatist ideologies, the 1980s saw the emergence of religiously motivated terrorism; while right-wing terrorism was prevalent during the 1990s, the most recent decade has been dominated by single issue attacks.
A key question then is what distinguishes terrorism hot spots from areas less frequently or never targeted by terrorists? Some researchers (LaFree and Dugan, 2004; Clarke and Neumann, 2006) have suggested that terrorism functions similarly to ordinary crime. The results from our descriptive analysis of terrorism hot spots appear to support this conclusion. Additionally, our research illustrates that those counties that experience terrorist attacks have a higher ordinary crime rate than counties that do not experience terrorist attacks. A long history of research has documented a strong relationship between ecological variables and ordinary crime. Specifically, drawing upon the preeminent theoretical work of Shaw and McKay and contemporary extensions of their ecological theory, we examined whether differences in socio-economic status, residential stability, and population heterogeneity distinguished counties that experienced terrorism.
However, the findings of the models predicting the likelihood of terrorist attacks at the county level also challenge traditional ecological theory and community-level empirical research of ordinary crime. First, consistent with the ordinary crime literature, areas characterized by residential stability are also buffered from terrorist attacks. Residentially stable areas may benefit from stronger social ties and consistent norms. But second, and counter to traditional ecological theory, whereas socioeconomic status and specifically concentrated disadvantage evidences a robust positive relationship with ordinary crime, the results of this analysis reveal that terrorist attacks are less likely to occur in areas characterized by concentrated disadvantage. While the finding that concentrated disadvantage is negatively related to terrorism is counter to what we would expect based on theories drawn from the ordinary crime literature, this finding is consistent with other research examining community correlates of terrorism (Krueger and Maleckova, 2003; for a review, see LaFree and Ackerman, 2009). Moreover, we must also note that whereas most of the research examining the relationship between concentrated disadvantage and ordinary crime is conducted at a neighborhood or city level, the current research was conducted at a higher level of aggregation which may account in part or in total for the disparate findings.
Finally, an interesting finding emerged from this research when examining the relationship between our indicators of population heterogeneity and terrorism. In support of contemporary research examining the effects of population heterogeneity on ordinary crime, the findings presented here demonstrate that the percentage of a population that is foreign-born in a county does not significantly influence the likelihood of terrorist attacks. Some have argued however that this variable does not adequately capture the idea of population heterogeneity as a community could have a high concentration of foreign-born people yet have minimal levels of heterogeneity (see Graif and Sampson, 2010). Therefore, we also examine the effect of language diversity as a measure of heterogeneity. Counter to recent empirical findings, language diversity evidences a strong and significant positive relationship with terrorist attacks and ordinary crime. Much more work needs to be done to fully understand the relationship between language diversity and terrorism and ordinary crime. In particular, in future research we plan to identify and isolate potential effects of specific language groups.