How does recalling experiences of conflict affect behaviour that aids or hinders post-conflict recovery
How to cite
Rigterink, Anouk; Sharp, Sam; Ayee, Tracy; Escande, Alice; Keyman, Ariana; Swanson, Sarah; Schomerus, Mareike. 2018. "How does recalling experiences of conflict affect behaviour that aids or hinders post-conflict recovery." AEA RCT Registry. AEARCTR-0003008. https://doi.org/10.1257/rct.3008-1.1
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
Numerous studies suggest that past experiences of violent conflict can affect deep determinants of behaviour (see (Bauer et al. 2016)). Perhaps contrary to intuition, the consensus among these studies is that past experiences of violent conflict promote pro-social behaviour. Two important questions remain regarding this literature: (1) is the relationship observed causal? As opposed to driven by reversed causality or attrition bias; (2) if so, why does this relationship exist?
We use an experimental set-up using priming to better establish causal relationships (so far only used by (Callen et al. 2014) for economic outcomes). The study design makes use of micro-narratives as a prime, which the respondents classify themselves along various dimensions. This means that respondents are asked to recall a significant experience and analyse it using the Sensemaker framework, providing insight into the analytical frameworks people use. Half of respondents (N=400) are primed to account a micro-narrative of their conflict experiences, half is primed to account a neutral narrative, prior to playing a set of behavioural games to measure deep determinants of behaviour. This will be followed by qualitative interviews.
Qualitative research and the use of self-classified micro-narratives will enable us to unpack the mechanisms behind the relationship between exposure to violence and behaviour. We will consider subsets of respondents for whom this relationship is especially strong or weak, and detect any patterns in the narratives these sub-groups tell.
We also reflect on the methodological lessons we have learned from using this integrated research design, including a consideration of how those at the receiving end of behavioural experiment experienced this type of research. Has this approach given us more rigorous and grounded insights—and can these insights help in developing better policies and programme approaches?
Intervention
Intervention is a priming treatment. EITHER:
a. The participant recounts a narrative of their experience during the Government of Uganda - Lord's Resistance Army conflict (treatment)
b. The participant recounts a neutral narrative (control)
Experimental design
Participants will be asked to participate in a series of consecutive exercises:
1. Priming treatment . EITHER:
a. The participant recounts a narrative of their experience during the GoU/LRA conflict (treatment)
b. The participant recounts a neutral narrative (control)
2. Self-signification. The participant classifies their narrative on a set of multi-dimensional scales which involves a series of questions that show the analytical framework which respondents use to make sense of experiences in their lives.
3. Behavioural games, to elicit preferences and pro- and anti-social behaviours.
4. Closed-ended survey questions, to assess the efficacy of the prime.
5. Qualitative interviews in the area affected by the conflict to establish how people describe the ongoing effect of conflict on their lives. These interviews examine the extent to which NGO programming continues to use a post-conflict lens that uses conflict recall and whether those elements that were largely considered part of the conflict dynamics at the time—such as neglect, exclusion, underdevelopment—still feature in people’s daily experience.
The main analysis will be a comparison of behaviour as measured by the games between treatment and control group.
Design details
Cited by
No citation data available for this registration yet.