Continuing with the next posting on the nineteenth lecture from Professor Michael Spagat’s Economics of Warfare course that he gives at Royal Holloway University. It is posted on his blog Wars, Numbers and Human Losses at: https://mikespagat.wordpress.com/
This lecture continues the discussion of terrorism, this time he is looking at a paper by Gassebner and Luechinger on terrorism (starts on page 13 of the lecture). This is similar to what was done for causes of war in a paper by Hegre and Sambanis that was presented in lecture 11:
So, the two authors ended up running a large number of regressions trying out many different combinations of variables and looking for those that are consistently significant. They used three different terrorism databases, resulting in them testing 18 different variables to each of the three different databases and looking at locations, victims and perpetrators (see slides 14 and 15). This gets a little involved and you are probably not going to sort it all out unless you read their paper (which costs $40 to order a copy…I did not).
What is particularly interesting to me is that the same variable has different values depending on what database is used. The values are of Coef. (median coefficient estimate), CDF (cumulative distribution function) and % sig. (the percent the estimate was statistically signicant). For example “Ethnic tensions” has values of -0.0007, 0.544 and 18.4 in the Iterate database, has values of 0.040, 0.911 and 63.1 in the GTD databases, and has values of 0.011, 0.627 and 15.2 in the MIPT database. Not sure what this really means (see slide 14). I have never done any analysis using someone else’s data base. I have always collected by own data and analyzed it.
Anyhow, the results are (see slides 18 & 19).
- Strong police states with religious tensions seem to be favored targets of terrorists.
- Bigger, economically repressive and richer countries seem to attract terrorist attacks.
- “Law and order” seems to discourage terrorism (this does conflict with point 1 above).
- A more foreign portfolio investment seen positively associated with terrorism.
- Higher natural resource exports as associated with fewer attacks on a county’s citizens.
- Citizens from countries with a “youth bulge” are attacked relatively less and do not attack more often.
- Fewer telephones are associated with more attacks.
- Countries with centrist governments seem to export terrorists
Anyhow, not finished with this lecture yet. Will have one more posting to do. The link to his lecture is here: http://personal.rhul.ac.uk/uhte/014/Economics%20of%20Warfare/Lecture%2019.pdf