Working Papers

Survey Participation Effects in Conflict Research

with Alexander De Juan, R&R at Journal of Peace Research.

Academic and policy interest in post-war political opinion has increased tremendously. One unexpected consequence of this surge of survey research is a growing probability that individuals will be interviewed multiple times. However, if participating in one survey causes respondents to change their attitudes or behavior, their subsequent survey responses may be biased in comparison to the rest of the sample population. Our paper aims to investigate such “survey participation effects” in conflict contexts. We draw on original, population-based survey data collected in the Eastern Democratic Republic of Congo (DRC). Our primary explanatory variable is previous survey participation. A staggering 18 percent of respondents in our representative sample report that they have been interviewed before. Multivariate analyses demonstrate that their stated attitudes on social relations, political institutions, gender norms and war-time victimization differ substantively from the responses of first-time interviewees. Moreover, our analyses indicate that experienced respondents have specific response styles – in particular, a tendency to support extreme response options. Lastly, we show that previous survey experience could in principle bias certain standard models on the sociopolitical legacy of violence. These findings underscore the potential effects of field research on the societies under investigation and call for a more careful consideration and analysis of the impact of survey data collection in and on conflict contexts.

 

Female Leaders, Regime Type and Public Goods Provision

with Anja Osei

Our article starts with the observation that the share of women in parliaments has increased steadily over the last decades worldwide. Apart from the descriptive representation of women, do more women in parliament substantively affect public policy more generally? And if so how? Drawing on a rich literature that suggests women to be more prosocial than men on average, we argue that a higher share of women in parliaments increases the salience of other-regarding preferences in the legislative arena. By extension, this manifests in higher degrees of public goods policies. We refine our argument suggesting that this effect is conditional on a state’s regime type. We test our arguments with a mixed-method design. First, using time-series cross-sectional data we find that up from a share of 10%, female MPs make a substantive difference in increasing public (as opposed to private) goods provision. We also find that this effect depends on regime type and only works in democracies. Second, we use coarsened exact matching to select two cases along a most-similar systems design. The quantitative analysis lends strong support to our argument and thereby contributes to filling a gap in the literature on the substantive impact of women in politics.

 

Conflict-Related Sexual Violence, Misreporting and Social Resilience: Evidence from a List Experiment in the DR Congo

with Summer Lindsey and Angelica Becerra

Misreporting to sensitive questions including conflict-related sexual violence (CRSV) is common in surveys. To address this problem, we administered a list experiment (aka item count technique) to 1,000 respondents in eastern Congo. First, we show that the estimated prevalence of CRSV changes from 6% in a direct question to 12% in the list experiment, suggesting that 6% of the respondents have misreported CSRV exposure. Second, we compare the effect of the direct question and the list experiment in a multivariate regression framework on several measures of social integration. We find no evidence that CRSV is related to social exclusion as suggested by previous qualitative studies. On the contrary, the list experiment reveals that CRSV-affected households are more socially active and engaged in their community, a finding that resonates with recent quantitative studies. We explore some potential mechanism and find that misreporting is related to anticipated stigmatization, social norms related to conformity as well as perceived state support. The paper contributes to a major gap in research on sexual violence by comparing direct and indirect techniques to measure CRSV.

 

Rape by Armed Groups, Social Stigmatization and Support Interventions: Evidence from Eastern Congo

with Summer Lindsey

Qualitative studies have shown that victims of rape by armed groups and their families experience social stigmatization. We build on this important work and provide new theory on the societal conditions under which stigma intensifies and decreases. Drawing on an original population-based survey experiment in the eastern Democratic Republic of Congo (DRC), we document the effects of rape by armed groups on stigma. On average victims and their families experience higher levels of stigma compared to unaffected families. We explicate a social theory of stigma to better understand the nature of these effects, finding that the effects of rape are dependent on community norms. Furthermore, our analysis shows that when rape-affected families participate in support interventions designed to address the social nature of stigma, stigmatization becomes indistinguishable from unaffected households. Our article builds on the findings from qualitative studies and significantly expands our knowledge by providing a refined theory and representative data on the prevalence of conflict-related sexual violence, its effect on stigma and the role of interventions in mitigating the negative effects.