Empirical Approach

Understanding the micro-level impacts of civil wars requires detailed, granular data at both individual and community levels. Existing data sources (e.g., Demographic Health Surveys, Afrobarometer, LAPOP) often lack the necessary depth to capture the nuanced effects of civil war exposure, including shifts in roles, responsibilities, and social dynamics. To address this gap, WarEffects employs an innovative research design and data collection strategy tailored to uncover these relationships.

Case selection

WarEffects focuses on three countries with diverse post-conflict experiences: Colombia, DR Congo, and Sri Lanka. These cases allow us to examine how civil wars influence community cohesion, gender dynamics, and local politics across different contexts. The selection reflects variation in:

• World region: Representing South America, Central Africa, and South Asia

• Conflict dimensions: Spanning ideological, ethnic, and religious conflicts

• Gender dynamics in armed groups: Differing levels of women's participation in combat roles

• Time since conflict: Ranging from ongoing to over a decade post-conflict

This comparative approach offers robust tests of the theoretical framework, ensuring insights are generalizable across diverse conflict settings.

Case Selection Strategy

To empirically assess the research questions and hypotheses, WarEffects will collect new individual-level and community-level quantitative survey data in three countries: Colombia, DR Congo, and Sri Lanka. While the project’s main goal is to explore the micro-level effects of civil wars and women’s empowerment in a representative population-based sample within each country, it also aims to compare between them, and generalize beyond these three country cases. The case selection strategy reflects this endeavour by leveraging cross-case variation along four key categories: world region, conflict dimensions, prevalence of sexual violence by armed groups, the extent of women’s participation in armed groups, and the time since the conflict ended. Replicating similar findings under very different country-level conditions constitutes a particularly hard test for the theoretical framework under investigation. 

Phased Research Design

Phase 1: Exploratory qualitative research

The first phase involves careful study of the historical, geographic, and sociopolitical contexts of each country. Using desk-based research and fieldwork, we will analyze variations in conflict dynamics and local discourses on gender, community cohesion, and political change. These insights will inform the design of surveys, ensuring they are culturally and contextually appropriate while maintaining comparability across cases.

Phase 2: Quantitative Data Collection

In each country, we plan to implement two types of surveys:

• Individual-level surveys: Gathering data on personal exposure to violence, roles in households, communities, and local politics, as well as attitudes toward gender and governance.

• Village-level surveys: Capturing community-level variables, including conflict events, aid programs, and local governance structures.

These surveys integrate experimental methods, such as list experiments to enhance privacy and accuracy and conjoint experiments to mitigate biases when measuring attitudes. This approach ensures reliable, actionable insights.

Phase 3: Follow-Up Fieldwork and Ground-Truthing

The third phase involves qualitative methods to contextualize and validate survey findings. Using semi-structured interviews and focus group discussions, we will explore how changes in community cohesion, gender dynamics, and local politics interact across different settings. Additionally, this phase emphasizes sharing results with local communities and stakeholders to foster engagement and practical application.