Preventing Islamic radicalization in Mozambique: through faith or employment?

This pre-analysis plan outlines the hypotheses to be tested in the evaluation of an initiative against Muslim radicalization done by the main Muslim authority in Mozambique. This initiative will be conducted in Mosques in the Province of Cabo Delgado at a time where a series of deadly attacks in the region was being attributed to Muslim radicals. In response to these events, the Mozambican Islamic Council (CISLAMO) decided to test two different types of interventions to dissuade the Muslim youth to join the radicals. A religious sensitization meeting where the authorities discuss the Muslim Law and Tradition to show how the radicals are not defending the Muslim Faith. And a professional training session that helps to increase youth employability. In this study we conduct a randomized controlled trial to evaluate the impact of these different programs.

2019-01-08

2019-01-10

We will collect data from three different sources:
1. “Joy-of-destruction” game with multiple partners.
This game follows the design developed by Abbink and Herrmann (2009). In the game, two players are endowed with 10 monetary units (MU) and simultaneously decide whether or not to destroy 5 MU of the other player at an own cost of 1 MU. We will ask each participant to play the exact same game with 4 different players: a Mozambican Muslim (same religion + same nationality); a Christian Mozambican (different religion + same nationality); a Mozambican Public official (local government) and a foreigner living outside of Mozambique (different nationality). The script of the game is shown in Annex A.
2. List Experiment.
We will conduct a list experiment across the three randomized comparison groups. So that we have a direct measure of the impact on reported support of extremist ideology, without many of the set-backs of self-reported questions. The script of the list experiment is in Annex B.
3. Survey.
The survey questionnaire collects standard demographic characteristics of participants, such as education and ethnic group; socio-economic variables, such as occupation, income and assets; satisfaction and expectations about the economic situation of the country and one-self and knowledge about recent discovery of natural resources in the region. Additionally, we collect data on social capital, trust in institutions and conflict. Annex C shows the complete survey.

In collaboration with our partners, we will select a group of 210 male individuals with ages between 18 and 30, from the Muslim community in the capital of the province of Cabo Delgado. The selection process of participants will be done by CISLAMO, under our supervision. Afterwards, we will randomly assign each of the 210 individuals to one of three groups:
a) Religious workshop (70 participants);
b) Professional training workshop (70 participants);
c) Control group (70 participants).

Done in the office by a computer.

Individual.

No

NA

210 in the main sample of interest.

70 individuals per group.