The War on Terror: A Data-Driven Learning Guide
Goal
The goal of this exercise is to consider the impact of the War on Terror in the global arena. We will also learn more about terrorism, generally. Crosstabulations and frequency tables will be used.
Concept
Terrorism can be defined as shocking acts of violence undertaken for the purpose of altering a government's policies. The violence usually targets those who do not actually have the personal authority to alter or enforce governmental policy.
Terrorist acts, because they are perpetrated on regular citizens, are meant to create fear or anxiety among the people, who then put pressure on a government to change its policies toward the terror organization.
The War on Terror is part of the United States' response to the September 11th attacks on the World Trade Center. The official objectives of the 2001 War on Terrorism are to counter terrorist threats, prevent terrorist acts and curb the influence of terrorist organizations.
Examples of research questions on the War on Terror include:
- Have terrorist attacks increased or decreased since the War on Terror began?
- Which regions of the world experience the most terrorist attacks?
- What are common targets of terrorist attacks?
- How do governments respond to terrorist attacks?
- Who is responsible for terrorist attacks?
Data for this exercise come from the Global Terrorism Database II (1998-2004). The goal of the Global Terrorism Database II (GTD2) was to create a comprehensive and sound dataset on global terrorism that can be used to derive methodologically robust insights into the phenomenon of terrorism and how to counter it. Information in the GTD2 was drawn entirely from publicly available, open-source materials. The criteria for incident inclusion and the coding scheme used in GTD2 were developed by a Database Development Committee, which consisted of recognized experts in terrorism and data collection. In order to be considered a "terrorist incident" the event had to have been committed by non-state actors had to have been violent, and intentional. In addition the act must have met two of the following three criteria: (1) The act must have been aimed at attaining a political, economic, religious, or social goal. In terms of economic goals, the exclusive pursuit of profit did not satisfy this criterion. (2) There must have been evidence of an intention to coerce, intimidate, or convey some other message to a larger audience (or audiences) than the immediate victims. (3) The action must have been outside the context of legitimate warfare activities, i.e. the act must have been outside the parameters permitted by international humanitarian law (particularly the admonition against deliberately targeting civilians or noncombatants). Data in this collection contain 7,154 events. The main variable categories presented in these data include: identification numbers, incident date, incident location, incident information, attack information, target information, perpetrator information, perpetrator statistics, perpetrator claim of responsibility, weapon information, casualty information, consequences, hostage/kidnapping, additional information, and source information.
Variables used in this exercise include:
- Region of Attack (REGION)
- Year of Attack (YEAR)
- Goal of attack: Political, Economic, Religious, or Social Goal (CRITERION1)
- Intent of the Attack: Coerce, Intimidate or Publicize (CRITERION2)
- Outside International Humanitarian Law? (CRITERION3)
- Psychosocial Consequences (PSYCHOSOCIAL)
- Attack Type - Alternative Categorization (ATTACK_TYPE0)
- Target of Attack (TARGET_TYPE)
Events: Terrorist Acts or Not?
As described above, only certain events were included in the dataset. Consider CRITERION1 where researchers coded events to determine if they had a "political, economic, religious, or social goal." Coding choices were "Yes" or "No." What percentage of the events in this dataset did researchers determine met this criterion?
Now consider CRITERION2 where researchers coded events to determine if they had "intention to coerce, intimidate or publicize to larger audience(s)." Coding choices were "Yes" or "No." What percentage of the events in this dataset met this criterion?
Finally, look at CRITERION3 where researchers coded events to determine if they were "outside international humanitarian law." Coding choices were: "Yes" and "No." What percentage of events in this dataset met the third criterion?
Consequences of Terrorism
Next think about the psychosocial consequences of terrorist events. Consider the frequency distribution of PSYCHOSOCIAL coded as: "major," "minor," "moderate," "none," or "unknown." In which category did most of the events fall?
Targets
Now look at the targets of terrorist attacks (TARGET_TYPE). Targets are coded into 27 categories, as shown in the frequency distribution. Given the length of the list, it may be helpful to look at the bar chart. What is the most common civilian target in this dataset? What is the second most common civilian target?
War on Terror
One way to evaluate the impact of the War on Terror using this dataset is to look at the number of attacks in each YEAR. For ease of interpretation, we recoded that variable into POST2001 such that all years from 1998-2001 are coded as 0, and while the years 2002-2004 are coded as 1. According to this dataset, were there fewer or more events before the War on Terror?
Events around the World
Consider the variable REGION, which categorizes where events took place. The regions include: "Australasia and Oceania," "Central America and Caribbean," "Central Asia," "East Asia," "Eastern Europe," "Middle East and North Africa," "North America," "Russia and the Newly Independent States (NIS)," "South America," "South Asia," "Southeast Asia," "Sub-Saharan Africa," and "Western Europe. " Look at the crosstab of POST2001 and REGION. Did North America experience more or less events prior to the War on Terror? Which two regions experienced the most events overall? Which experienced the fewest? In which regions did terrorist incidents increase, proportionally?
Events are also characterized by ATTACK_TYPE0. The categories are: "armed assault," "assassination," "basic assault," "bombing/explosion," "facility / infrastructure attack," "hijacking," "hostage taking (barricade incident)," "hostage taking (kidnapping)," and "unknown." Look at the crosstab of ATTACK_TYPE0 by REGION. Which region had the most "bombing/explosion" events? Which region had the most "hostage taking (kidnapping)" incidents? Which region had the most "facility/infrastructure" events?
Think about your answers to the application questions before you click through to the interpretation guide for help in answering them.
What percentage of the events in this dataset had a "political, economic, religious, or social goal?"
What percentage of events had "intention to coerce, intimidate or publicize to larger audience(s)?"
What percentage of events in this dataset were "outside international humanitarian law?"
In which category of level of consequence did most of the events fall?
What is the most common civilian target in this dataset? What is the second most common civilian target?
According to this dataset, were there fewer or more events before the War on Terror?
Did North America experience fewer or more events after the War on Terror? Which two regions experienced the most events overall? Which regions experienced the fewest? In which regions did terrorist incidents increase, proportionally?
Which region had the most "bombing/explosion" events? Which region had the most "hostage taking (kidnapping)" incidents? Which region had the most "facility/infrastructure" events?
Interpretation
Reading the results:
- Frequency distribution tables (crosstabs) display percentages in each cell of the people who fall into the overlapping categories (if two variables are used) followed by the actual number of respondents who fall into the overlapping categories (cells). The coloring in the tables demonstrates how the observed numbers in each cell compares to the expected number if there were no association between the two variables. Paler colors represent lesser significance of the difference between observed values and expected values. The accompanying bar charts display the patterns visually as well.
- Weights (mathematical formulas) are often used to adjust the sample proportions, usually by race, sex, or age, to more closely match those of the general population. The analyses used in this guide did not use any weights, which may reduce the generalizability of the findings, but the resulting tables are accurate descriptions of the relationships found between these variables among these respondents.
The analyses show the following:
- 92.9% of the events were coded as having "political, economic, religious, or social goals," so 7.1% did not meet this criterion. 93.3% of the events were coded as having the "intention to coerce, intimidate or publicize to larger audience(s)," so 6.7% did not meet this criterion. Finally 96.6% were determined to be "outside international humanitarian law," so 3.4% did not meet this criterion. Recall that in order to be included in the dataset, an event had to meet two out of three conditions.
- Most of the events were coded as having "Minor" psychosocial consequences (83%).
- In this dataset, "private citizens and property" was the most common civilian target (30.8% or 2203 events). The second most common civilian target was "business" with 8.9% or 637 events.
- According to this dataset, there has been a decline in the number of terror attacks per year since the onset of the War on Terror. From 1998-2001 there were 4,409 attacks or about 1,100 per year. From 2002-2004 there were 2,745 attacks or about 915 per year.
- Similarly, there has been a decrease in terror attacks in North America since 2002 (about 17 per year between 1998 and 2002 and about 8 per year between 2002 and 2004). Overall, the Middle East and North Africa (1,617) and South Asia (1,580) experienced the most events, as well as the largest increases in terrorist incidents as a proportion of all incidences. Before 2002, 19.9% of terrorist attacks took place in the Middle East-North Africa, and 18.1% in South Asia. These percentages increased to 26.9% and 28.5% respectively during the 2002-2004 time period.
- Throughout the entire time period encompassed by the study (1998-2004), Australasia and Oceania (19 attacks) and Central America and Caribbean (27 attacks) experienced the fewest terrorist events of any type. The highest percentage of "bombing/explosion" attacks took place in the Middle East and North Africa (23.8%), followed closely by South Asia with 23.5%. South America experienced the greatest number of "hostage taking (kidnapping)" incidents. 29.9% of kidnappings happened there. Western Europe had the most "facility/infrastructure" events (162, or 27.8%).
Summary
In this exercise we learned that not all events met each of the three criteria for terrorist attacks. We also saw that the most common psychosocial consequence of the attacks was "minor." Private citizens and property were the most common targets of attacks. Since the launch of the War on Terror in 2002, there have been fewer terrorist events overall, though two regions, the Middle East and North Africa, and South Asia, saw an increase proportionally in incidences of terrorism.
Although the goals of terrorist acts are to influence government policies by harming innocent citizens, this dataset is unable to address whether or not governments adapted policies based on events. Researchers may wish to consider government responses to events, as well as evaluate the long term impact of the War on Terror. Further research may try to understand better the groups that execute terrorist attacks in order to assess the outcomes of such attacks fully.
CITATION: Inter-university Consortium for Political and Social Research. The War on Terror: A Data-Driven Learning Guide. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2009-04-16. Doi: https://doi.org/10.3886/terrorism
This work is licensed under a Creative Commons Attribution-Noncommercial 3.0 United States License.