The Legacy of Political Violence, Wellbeing, and Ethnicity: An Experimental Analysis of Cote d'Ivoire and Uganda

How to cite

Doces, John; Adou, Donatien. 2022. "The Legacy of Political Violence, Wellbeing, and Ethnicity: An Experimental Analysis of Cote d'Ivoire and Uganda." AEA RCT Registry. AEARCTR-0010213. https://doi.org/10.1257/rct.10213-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

John Doces* , Donatien Adou

* Primary investigator

Registry ID
AEARCTR-0010213
DOI
https://doi.org/10.1257/rct.10213-1.0
First registered
Last updated
Status
In development
Trial dates
2022-10-17 – 2022-11-08
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/10213

Abstract

Across the world the legacy of political violence persists for people often affecting their lives in adverse ways. Past events of political violence often associated with elections continue to threaten people affecting their wellbeing and identities. Existing scholarship shows that the legacy of political violence can inversely affect people's wellbeing and also their identities pushing people to feel less satisfied and to identity more closely with their ethnic group. By reducing wellbeing and causing people to identity more with their ethnic group the legacy of political violence can in turn increase the probability of further political violence. We test these ideas in Africa focusing on a unique comparative analysis in West and East Africa both of which are regions that have experienced past political violence and thus carry its legacy. We specifically examine the legacy of political violence, wellbeing, and ethnic identities. We conduct an experimental analysis treating randomly selected respondents in Cote d'Ivoire and Uganda and treating some with a statement about the legacy of political violence with a control group assigned no statement. We have several claims: (i) respondents treated with legacy of political violence will have lower levels of wellbeing than those not treated; (ii) the treatment effect on wellbeing will be relatively larger for women than men; (iii) respondents treated with political violence will have more ethnic than national identities. As an extension and for purposes of comparison we will test these claims in a nationally representative survey (filled by yougov) of Americans to see if the results are consistent.

Intervention

Interventions will include statements about past political violence tailored to each country's recent past (Cote d'Ivoire, Uganda, and the United States). Each respondent that receives the intervention will read a statement summarizing past events of political violence along with a picture emphasizing the violence.

Experimental design

Experiment will be in the field. Random assignment of treatment and control groups will be done using pre-determined randomized ordering of surveys per enumerator. Respondents will not be compensated.

Design details

Randomization method
Surveys will be randomly assigned using a random number generator.
Randomization unit
Treatment will be randomized at individual level.
Clusters
This is not a cluster analysis.
Observations
In Cote d'Ivoire and Uganda we will collect 320 observations in each country. In the U.S. the survey will collect approximately 1,000 observations.
Arms
Treatment group will have 320 observations and control group 320.
Primary outcomes
Key outcome variables include wellbeing (measured by Cantril Index) and ethnic versus national identity as used by Afrobarometer.

Cited by

No citation data available for this registration yet.

Keywords