X Close

Military Affairs & Counter-terrorism Society (MACT)

Home

Menu

Archive for August, 2018

Student Research Showcase #2

By uctzaes, on 10 August 2018

Over the week we met up with Alexandre Bish at the Printroom Café so that he could introduce us to the research he is currently conducting as part of his dissertation for the the MSc course in Countering Organised Crime and Terrorism Studies at UCL. What follows is our discussion:

Hello Alex! Could you begin by telling us your current research endeavour as part of your dissertation and specify your research questions?

Sure thing.

As part of my dissertation, I am looking at what features a given location make it vulnerable to a terrorist attack. In other words, what are the key attributes of a location that make a terrorist chose that location over another when planning an attack. Assessing which locations are most vulnerable is key, as it will enable a better allocation of policing resources (especially considering budget constraints). It will also help target interventions that will reduce the ‘attractiveness’ of vulnerable locations.

Two seminal contributors to crime science, Clarke and Newman (2006) wrote a book entitled Outsmarting the Terrorists, in which they presented an untested framework for assessing the vulnerability of locations to terrorist attacks. They argued that vulnerable locations contain some (or all) of the following features: vulnerable locations are Exposed, Vital, Iconic, Legitimate, Destructible, Occupied, Near (to the terrorist’s base) and Easy (to attack). These indicators all come together in the wonderful acronym ‘EVIL DONE’.

A good measure of the EVIL DONE constructs is essential in order to test their utility. With a view to standardise the measure of the EVIL DONE constructs, Boba (2009) presented a scale with objective criteria to rate location vulnerability. My study uses (and tests) this scale to evaluate location vulnerability.

My first research question was to see how reliable the scale was. I asked three coders (to whom I am eternally grateful) to rate 115 attacks which would help me assess the scale’s inter-rater reliability (i.e. the level of agreement across raters). As Boba’s scale was quite specific, I had to create my own database of terrorist attacks. I found that testing suicide bombings in Israel during the Second Intifada was the best way for me to test the EVIL DONE for four main reasons:

  • The sample size of attacks was large enough to run inferential statistics.
  • Suicide bombings in Israel have good coverage in the media: I was able to find detailed information of attacks, crosscheck information, and compile photographs of locations where attacks took place (these often pictured the aftermath of attacks). This allowed for the effective identification of location features, which helped code most EVIL DONE variables.
  • The Israeli Security Agency released detailed information on suicide bombers (including information on where attacks were planned). With this information, coders had enough information to rate the ‘Near’ indicator (which assess the proximity to terrorist bases).
  • Israel has a good coverage of Google Maps and locations were found on Google Street View. Screenshots of the Google Street View for each attack location were compiled to bring extra information and thus more accurate rating.

Once the three coders rated the 115 events, I used the average rating for each indicator across the three coders to conduct inferential statistics. I tested the construct validity of the EVIL DONE through factor analysis by evaluating which indicators clustered together. In other words, I wanted to see whether some constructs actually measured the same thing.

Finally, I looked at the utility of the EVIL DONE framework to see whether the EVIL DONE actually predicted vulnerability. I also looked at whether the EVIL DONE predicted the seriousness of an attack. To do this, I ran regressions with measures of seriousness and vulnerability as dependent variables.

That all sounds great! What is your research so far suggesting are the most significant factors in the EVIL DONE model? Would you be able to rank them from most important to least? Could you tell us a bit more about your findings?

I cannot go into too much detail about my findings. Fingers crossed, if I manage to put a paper together that is good enough for publication, you will manage to read all about them! This dissertation is a first step in what I hope will make an interesting side-project of mine. Ultimately, the goal would be to create a convincing model for terrorist target selection, and map the model to allocate policing resources and tailored situational interventions.

Finally, what are the two readings that informed your study the most?

I guess the two readings that have informed this study the most are those that I have already briefly mentioned. Outsmarting the Terrorists by Clarke and Newman (2006) and Responding To Terrorism Through Situational Crime Prevention edited by Freilich and Newman (2009) in which Boba presents her scale in a chapter entitled EVIL DONE.