The independent resource on global security

Measuring conflict exposure in micro-level surveys

While violent conflict is a key obstacle to human development, the impact of conflict on people varies greatly. Individual conflict experiences may vary by gender, political view, socio-economic status or mere bad luck. Socio-economic research on conflict has demonstrated that the best route to overcoming legacies of a conflict depends on the circumstances of the conflict. Successful peacebuilding and reconstruction policies must therefore build on specific local and individual conflict legacies, considering how people were affected by war and violence.

The Conflict Survey Sourcebook, published in August 2013 by the Word Bank, serves as a basis for developing techniques for conducting socio-economic surveys in conflict-affected and insecure contexts.

Patricia Justino, Andrew Tedesco, Philip Verwimp and I were motivated to write the Sourcebook because we felt that we were hitting a barrier in our research. There were certain questions we couldn’t answer because the data sets that were available to us were incomplete. We realized there were detrimental gaps in the data being collected—everywhere we looked we had partial answers.

Explaining conflict surveys

Underlying all surveys in conflict-affected areas is the assumption that every aspect of life is deeply impacted by war, violence and insecurity. Yet most surveys implemented in such settings do not ask about war—or if they do, they do so inconsistently. Yet if you ask about displacement but not looting of assets, then you will get a biased understanding of the conflict dynamics and of a conflict’s impact on human development.

You can capture conflict experience for individuals through direct and indirect effects.  First, there are direct impacts on life experience: people are killed or injured by conflict, they get forcefully displaced, men are drafted into the army and so on. Many of these impacts can be captured by standard surveys. However, while a standard survey may show that someone died recently, it may not record that the cause of death was war-related. Failing to include these conflict-specific codes leaves significant holes in the data.

Accounting for indirect consequences of conflict

Facts, events and choices are not the only relevant data in conflict-affected contexts. Indirect and abstract things such as fears, perceptions and expectations are also key features of conflict. Something direct could be a family member being killed as a result of military action, or an attack on a village, but an indirect consequence could be the presence in that village of a lot of people who are internally displaced from the war zone.

In this sense, your life is not directly affected by conflict but the presence of the displaced people who are willing to work for very little because they’re destitute might change the local labour market, or make local residents feel threatened. That then needs to be captured in the survey in order to understand the lives of people. Worries, suspicions, and fears are in fact a major burden on people and a large driver of unhappiness and mental ill health. Hence, asking about the indirect effects of conflict is also important.

Interviewing conflict-affected families in Kyrgyzstan

Together with a large team, I am currently conducting a survey in conflict-affected Kyrgyzstan, through SIPRI’s  Life In Kyrgyzstan project. In my experience with data and surveys in conflict zones, there are several key questions to consider during the design phase:

  • What should be the focus of the discussion in the short time available with conflict-affected families?
  • How can questions be posed in a way that respects the traumatic nature of the content you are asking them to share?
  • How valid are these answers as time passes since the event (has society’s narrative tainted the participant’s personal narrative?)


While hard facts are important, we also wanted to capture how people feel about more personal questions such as ‘Are you scared walking at night?’ or ‘Do you fear for the safety of your family members?’

There could be many reasons why people are scared but the answers may reveal patterns—for example, that people who were internally displaced are more scared than other people elsewhere, perhaps because of their extreme experience of displacement. Or perhaps families are worried about their daughters walking alone to school in conflict-affected areas, thus reducing the schooling of girls in a conflict-affected area. If we ask the right questions, these are the strands of arguments we can start to uncover.

Providing useful tools for practitioners and researchers

Surveys done by practitioners—for example, people working for aid agencies who are carrying out needs assessments—may be asking different questions than a survey done by a PhD student doing field research.

In either case, capturing how people experience conflict dynamics requires increased capacity on the part of practitioners and researchers to identify how violent conflict affects individuals, households and communities along key social and economic dimensions. We hope that our Sourcebook provides a useful tool to help both practitioners and researchers systematically address all of the conflict-related factors relevant to their studies.


This blog post is published as part of a collaborative partnership between SIPRI and Economists for Peace and Security (EPS).