Chapter 1: | The Challenge of Global Terrorism |
This is a limited free preview of this book. Please buy full access.
Third, the analysis extends through December 2004. As a result, in addition to examining the effects of the invasions of Afghanistan in late 2001 and Iraq in March 2003, it looks at the inflaming effects of the capture of Saddam Hussein in December 2003 and the release of photos of alleged prisoner abuse from Abu Ghraib in April 2004.
Fourth, to further test the escalation hypothesis, I use new variables to measure potential escalation in transnational terrorist activity. In addition to examining moving averages of quarterly counts of incidents and of lethal incidents, I examine averages of quarterly days of transnational terrorist activity to see if there has been an increase in the number of operative days of transnational terrorist activity. This measure may be important since those who engage in terrorism seek publicity and the number of operative days may be a good proxy for media coverage. (The assumption here is that when terrorist incidents are mounted on more days, they are more likely to be covered more days in the media.) Also, on the basis that escalation implies a widening of activity, I examine if attacks are more dispersed; that is, if they occur in more places (countries) and if the percentage share of incidents in the Middle East and in predominantly Muslim countries is increased.
Fifth, on the basis that escalation could be seen to mean a widening of a conflict in terms of the targets, I examine whether there has been a change in the makeup of victims of attacks. Specifically, I test the extent to which victims of transnational terrorist attacks, since the onset of the War on Terrorism, are more likely to be multinational (i.e., include more than one nationality or to be members of multinational companies or multinational organizations such as the United Nations or international nongovernmental organizations).36
Finally, I provide comparisons between the ITERATE and RAND databases and between unique and overlapping incidents in the data sets. These comparisons are designed to shed light on the composition of each database and may be helpful in future research.