Does Higher Education Decrease Support for Terrorism?

Benmelech, Berrebi, and Klor (2012) use data on Palestinian suicide terrorists from 2000 to 2006 to show that human capital (education and experience) is an important factor for selecting terrorists for suicide tasks. Caruso and Schneider (2013) offer a different perspective on the tactic of Al Qaeda-inspired terrorist groups: using contest theory they show that terrorist groups compete with each other by increasing the number of attacks and their brutality.

For an interesting exception see Caruso and Gavrilova (2012) who argue that the level of education affects the sense of frustration and grievances among younger population. On the link between education and suicide terrorism see Azam (2012).

PEW http://www.pewglobal.org/category/datasets/ (Pew Global Attitudes Project: Spring 2007 Survey).

We excluded the Egypt – Egypt pair.

A specific case was the question regarding the European Union since it is not a country, though it can be considered a power. For the purpose of calculating the GDP, population and civil liberties, we calculated averages of countries that we assigned to the group EU (Germany, France, Belgium, UK, Spain, Italy, Luxemburg and Netherlands). We selected these countries as the oldest and leading members of the EU. This choice is in line with the collection of data on terrorist incidents against these countries in the period from 2004 to 2008.

For exact wording of the questions and details about the data check authors, 2011.

World Development Indicators http://data.worldbank.org/data-catalog/world-development-indicators.

In the NCTC Worldwide Incidents Tracking System (WITS) a terrorist incident is defined as an incident “in which subnational or clandestine groups or individuals deliberately or recklessly attacked civilians or noncombatants (including military personnel and assets outside war zones and war-like settings)” (The Worldwide Incidents Tracking System).

The fact that the values of the dependant variable range from 0 to 23 per pair raises problems of overdispersion and the test for overdispersion in our sample shows that it is significant V(y|x)=E(y|x)+a*{E(y|x)^2}.

The Poisson distribution assumes that the mean is equal to variance.

Our analysis of gender and age showed no significant effect. See footnote n. 16.

Figures A4–A6 show the correlations between the number of terrorist attacks and the shares of the critical group by levels of education.

[exp 39.36*0.018)–1]*100.

For the educational attainment of terrorists see Benmelech and Berrebi (2007), Krueger and Malečková (2003).

We also estimate the model separately for males and females and in both cases find similar results. These estimations suffer from omitted variable problems. (The share of the highly educated females and the share of the highly educated males in the critical group are highly correlated variables. Including both of these variables in the model will cause a problem of multicolinearaity between the variables.) We therefore do not include these estimations in the paper, but the result suggests that there is no difference in respect to gender.

[(0.018*6.773)*100].

p=[1–(0.00563241)]*[(5.2567*(1–5.2567)/(12315/(16)]/[(1–0.9385)*(18.92)]. For the definition and the variables from the equation (5) see Aydemir and Borjas (2010, 10).

[(0.018*6.604)*100].

However, Esteban and Ray (2008) argue that economic inequality may lead to a better division of labor among the rich, funding terrorism, and the poor providing the pool of recruits. Similarly, when educational inequality is high, the educated people may provide the “ideological” output while the non-educated people supply the “manpower;” under these conditions, the risk of terrorist attacks is higher.

See (http://www.nctc.gov/).

Analysis of WITS Impact on Scholarly Work on Terrorism (Krueger, Laitin, Shapiro and Stanišić, 2011, unpublished manuscript).

It would be interesting to know what type of tertiary is relevant. Gambetta and Hertog (2009) show that individuals with an engineering education are overrepresented among violent Islamists. Unfortunately, the survey data we use do not provide information about the type of tertiary education of the respondents.

In a different context, Azam and Thelen (2008) argue that foreign aid reduces terrorist attacks by recipient countries, as does the recipient country’s level of education. Claiming that foreign aid helps the receiving government fight terrorism, they suggest that the correlation between the level of education of the individual terrorists and their activism is irrelevant from the donor’s point of view, as the local government will adjust its level of repression optimally as a function of the impact of education.