2/8 Starting from the ground-breaking work by @paulgill_ucl @Drjohnhorgan and @pjdeckert in their 2013 #JOFS article ‘Bombing Alone’, we built a dataset on #loneactor attack planning and preparation, covering 55 cases in #NorthAmerica and #Europe, '78-'15http://onlinelibrary.wiley.com/doi/10.1111/1556-4029.12312/full …
-
Show this thread
-
3/8 In the first of 3 articles,
@paulgill_ucl@noemie_bouhana, Edwin Bakker and I use this dataset to show 1) most lone actors not highly lethal or stealthy, 2) they are generally poor at OpSec, 3) leak motives and intentions, 4) do so months in advance.http://onlinelibrary.wiley.com/doi/10.1111/1556-4029.13676/full …2 replies 9 retweets 5 likesShow this thread -
"most lone actors not stealthy", "leak motives and plans" sounds like selection bias. Your sample only included those who were caught and were therefore less competent in keeping their plans secret.
1 reply 0 retweets 0 likes -
Also included many (probably a majority) successfully executed attacks in the analysis. They are all there in a table.
1 reply 0 retweets 1 like -
I still think there is a general problem of sampling in that you can only profile those who are caught (even if it's only after they carry out their attack). Even so, it was an interesting study, so thanks.
2 replies 0 retweets 0 likes -
Sorry this is an idiotic point though. If we sample those who attack and those who get caught, who is left out of the study? Read the thing, it's open access.
2 replies 0 retweets 4 likes -
No need to be rude. My point is that while the study is useful, it is limited to those networks that are known. As you know, the true extent of networks are rarely found, and many attacks go unsolved. A list of those in Turkey alone can be found here: https://www.researchgate.net/publication/221246952_Prediction_of_past_unsolved_terrorist_attacks …
1 reply 0 retweets 0 likes -
You didn't say it was useful initially though, you criticised years of work without reading or engaging with it which, in my opinion, is rude. The study you cite is of terrorist groups not lone actors and turkey was outside of our geographical inclusion criteria.
1 reply 0 retweets 1 like -
My point wasn't intended as criticism of your paper specifically, but of selection bias in terror research generally. Statistical modelling of terror attacks will always suffer from the fact that the amount of detail available is proportional to the incompetence of the attacker.
1 reply 0 retweets 0 likes -
But you're presuming they don't want to get caught and have their identities known in the end. Most do! Lone actors rarely do multiple attacks across a period of time. The dynamics with these guys is very different than your traditional group-based operatives.
1 reply 0 retweets 3 likes
Okay, fair enough, that's an interesting point.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.