Terrorism, Instability, and Democracy in Asia and Africa

Chapter 4: Methods Used to Investigate the Causes of Terrorism in Asia and Africa

chapter 4 Methods Used to Investigate the Causes of Terrorism in Asia and Africa T he main purpose driving our research is to examine four factors—civilization, democracy, poverty, and regime stability— and their potential impact on international and domestic terrorism. As we have previously stated, we believe such an inclusive examination of terrorism is novel. Further, no study that we have come across examines domestic terrorism in multiple states. There is a small but growing research body that examines the impact of democracy on terrorism, but we feel research on terrorism is still in its embryonic stages. The book is divided into two distinct research sections. The first section is a quantitative examination of the four factors of civilization, democracy , poverty, and regime stability and their effect on both international and domestic terrorism. Except for regime interruptions, multiple variables were used to measure civilization, democracy, and poverty. The second section is broken down into distinct chapters that examine regional case studies. Our selection of relevant case studies was largely driven by our statistical findings. States experiencing high levels of domestic and/or international terrorism were automatically included as cases for examination. Similarly, some of the states experiencing low or no terrorism were used to compare and contrast with states experiencing high levels of terrorism. The examination of our regional cases is also conducted through the lenses of civilizational, democratic, poverty, and regime instability factors. We decided early on in this project to present our statistical findings as graphical, bivariate scatter plots using the SPSS statistical program. There are several reasons for our decision. First, while the Memorial Institute for the Prevention of Terrorism–Terrorism Knowledge Base (MIPTTKB ) contains twenty-nine years of data that we can use in our current 40 study on international terrorism, there are currently only six years of data available on domestic terrorism. Because we wanted to compare findings on international and domestic terrorism, and because six years of domestic terrorism data do not allow for enough degrees of freedom for an advanced regression technique, we decided that a graphical, frequency representation of our results was the best course of action. Second, this is a first effort at such a wide examination of international and domestic terrorism, and we wanted to make the results accessible to scholars, students, and policymakers. We place special emphasis in our goal of producing a work that will benefit policymakers. Third, graphical representations of cases allow for greater and easier interpretation of our results. Outliers are easier to pick up, and trends that might exist graphically, but that are harder to discern statistically, can be more readily examined and expounded upon. It is our hope that this book will provide a basis for further and more detailed inquiries. Having said this, our techniques do allow for a certain degree of statistical sophistication. SPSS allows us to add a constant to the binomial equation and generate a rough r-squared statistic. This allows some interpretation of the strength of the relationship. In a general sense, an rsquared of .05 indicates that roughly 5 percent of the variance in the dependent variable, which will be either international or domestic terrorism, is explained by the independent variable, some measure of civilization, democracy , poverty, or regime instability. SPSS also allows us to generate a fit line. This gives us a graphical representation of the directionality of our relationship. By “directionality” we are referring to the positive or negative nature of the correlation. If the slope of the line is upward, then the relationship is positive. If the slope of the line is downward, then the relationship is negative. A positive relationship means that when more of whatever you are measuring, like democracy, is present in society, then there is a greater frequency of terrorism. A negative relationship means that when there is more of whatever you are measuring, such as democracy, there is a lower frequency of terrorism. Unfortunately, we were unable to generate a significance level, but higher r-squared statistics generally correlate with higher significance in the relationship. In order to generate points on our scatter plots, averages for most statistics were used. For the international variables, averages were generated over the twenty-nine years of data, and for the domestic data the average spanned the six years of available data. There is some missing data in our data set. Some missing data comes from simple missing data points in our Methodology to Investigate the Causes of Terrorism 41 [3.21.231...