Predicting Suicide Attacks

Predicting Suicide Attacks: Integrating Spatial, Temporal, and Social Features of Terrorist Attack Targets

Walter L. Perry
Claude Berrebi
Ryan Andrew Brown
John Hollywood
Amber Jaycocks
Parisa Roshan
Thomas Sullivan
Lisa Miyashiro
Copyright Date: 2013
Published by: RAND Corporation
Pages: 112
https://www.jstor.org/stable/10.7249/j.ctt2jc9d2
  • Cite this Item
  • Book Info
    Predicting Suicide Attacks
    Book Description:

    As part of an exploration of ways to predict what determines the targets of suicide attacks, RAND conducted a proof-of-principle analysis of whether adding sociocultural, political, economic, and demographic factors would enhance the predictive ability of a methodology that focused on geospatial features. This test case focused on terrorist bombing incidents in Israel, but the findings indicate that the methodology merits further exploration.

    eISBN: 978-0-8330-7803-2
    Subjects: Political Science, History

Table of Contents

  1. Front Matter
    (pp. i-ii)
  2. Preface
    (pp. iii-iv)
  3. Table of Contents
    (pp. v-vi)
  4. Figures
    (pp. vii-viii)
  5. Tables
    (pp. ix-x)
  6. Summary
    (pp. xi-xxii)

    The threat of suicide bombings in the United States and elsewhere prompted the Department of Homeland Security to commission the Naval Research Laboratory (NRL) to develop a method for predicting the determinants of suicide bombing attacks. As a test case, NRL chose to study suicide bombings in four Israeli cities: Jerusalem, Haifa, Tel Aviv, and Netanya. They focused on three terrorist groups: Hamas, Al-Aqsa Martyrs’ Brigade, and the Palestinian Islamic Jihad.

    NRL designed a two-part study aimed at discovering terrorist group target preferences in suicide terrorism. The first part focused on examining spatial preference patterns: how the different terrorist groups...

  7. Acknowledgments
    (pp. xxiii-xxiv)
  8. Abbreviations
    (pp. xxv-xxvi)
  9. CHAPTER ONE Introduction and Overview
    (pp. 1-4)

    The threat of suicide bombings in the United States and elsewhere prompted the Department of Homeland Security to commission the Naval Research Laboratory (NRL) to develop a method for predicting the determinants of suicide bombing attacks. NRL chose to study suicide bombings in four Israeli cities: Jerusalem, Haifa, Tel Aviv, and Netanya during and after the Second Intifada (1993–2006). In addition, NRL focused on three terrorist groups: Hamas, the Al-Aqsa Martyrs’ Brigade, and the Palestinian Islamic Jihad. Because of Israel’s long struggle to combat suicide bombings, it possesses a larger data set suitable for quantitative analysis than does the...

  10. CHAPTER TWO Quantitative Data and Methods
    (pp. 5-38)

    We took a three-stage approach to the quantitative analysis of sociocultural data to augment the geospatial prediction of suicide bombing targeting. First, we used socioeconomic, demographic, and political data at the neighborhood level to examine neighborhood characteristics associated with the likelihood of attack. We used principal components analysis (PCA) to reduce data dimensions and logistic regression to model the likelihood of attack by neighborhood (Abdi and Williams, 2010; Pearson, 1901). Second, we used classification and regression trees (CART) (Brieman et al., 1984; Gruenewald et al., 2006; Therneau, Atkinson, and Ripley, 2008) to model higher-order combinations of variables associated with high-risk...

  11. CHAPTER THREE Qualitative Analysis
    (pp. 39-52)

    The purpose of the qualitative methodology was to identify and codethemesreflecting target location types and attributes common to suicide bombing sites in Israel. We used the hybrid NRL-RAND model to identify areas at increased risk of attack, then used the themes from the qualitative analysis to identify specific locations at increased risk of attack. The intention was to make the problem more manageable by reducing the high-risk areas from broad city regions to lists of sites. Figure 3.1 shows a notional example of how we used qualitative analysis to do this for areas of Jerusalem that the NRL...

  12. CHAPTER FOUR Conclusions and Recommendations
    (pp. 53-58)

    This final chapter records the major findings from this research. The first section addresses findings associated with the quantitative analysis reported in Chapter Two, and the second addresses the qualitative analysis addressed in Chapter Three. We conclude with a set of recommendations that focus on further research.

    The purpose of the pilot study documented here was to establish the potential utility of sociocultural (as well as political, demographic, and socioeconomic) information in assisting with the spatial prediction of suicide bombing attacks, using Israel as a test case. The quantitative analysis established that socioeconomic, demographic, and political data not only have...

  13. APPENDIX A Sociocultural Precipitant Database
    (pp. 59-70)
  14. APPENDIX B Logistic Regression Output
    (pp. 71-76)
  15. About the Authors
    (pp. 77-82)
  16. Bibliography
    (pp. 83-85)