A latent variable approach to measuring wartime sexual violence
| Published date | 01 November 2020 |
| Date | 01 November 2020 |
| DOI | http://doi.org/10.1177/0022343320961147 |
| Subject Matter | Regular Article |
A latent variable approach to measuring
wartime sexual violence
Jule Kru
¨ger
Ragnhild Nordås
Department of Political Science, University of Michigan
Abstract
Conflict-related sexual violence is an international security problem and is sometimes used as a weapon of war. It is
also a complex and hard-to-observe phenomenon, constituting perhaps one of the most hidden forms of wartime
violence. Latent variable models (LVM) offer a promising avenue to account for differences in observed measures.
Three annual human rights sources report on the sexual violence practices of armed conflict actors around the world
since 1989 and were coded into ordinal indicators of conflict-year prevalence. Because information diverges signif-
icantly across these measures, we currently have a poor scientific understanding with regard to trends and patterns of
the problem. In this article, we use an LVM approach to leverage information across multiple indicators of wartime
sexual violence to estimate its true extent, to express uncertainty in the form of a credible interval, and to account for
temporal trends in the underlying data. We argue that a dynamic LVM parametrization constitutes the best fit in this
context. It outperforms a static latent variable model, as well as analysis of observed indicators. Based on our findings,
we argue that an LVM approach currently constitutes the best practice for this line of inquiry and conclude with
suggestions for future research.
Keywords
armed conflict, latent variable model, measurement, observational data, sexual violence, uncertainty
Introduction
Wartime sexualviolence can have a severe negative impact
on survivors and affected communities. Policy and advo-
cacycommunities have calledfor improved documentation
and analysis to curb abuses and mitigate their harm. We
have also seen an upsurge in systematic empirical work on
wartime sexual violence (Leiby, 2009; Roth, Guberek &
Hoover Green, 2011; Cohen, 2013; Cohen & Nordås,
2014, 2015; Chu & Braithwaite, 2018).
Despite growing scholarly attention to wartime sexual
violence,its empirical study is fraughtwith methodological
challenges.For a range of reasons, sexual violence could be
more difficult to observe and measure than other conflict-
related violations. Historical data is largely missing and, if
available, often contradictory, making it difficult to deter-
mine temporal trends. Even today, with international rec-
ognition in place, empirical data on victims, survivors, or
perpetratorsis extremely difficultto obtain (Roth,Guberek
& Hoover Green, 2011).
Analyses of what drives wartime sexual violence can
only be as good as the empirical basis. Without taking
into account underlying data uncertainty, policy
resources will probably be spent in inefficient ways, sci-
entific conclusions might be inaccurate, and empirical
findings from otherwise carefully conducted studies
could be misinterpreted or misappropriated. More accu-
rate data and better understanding of data uncertainty
will equip policymakers and practitioners with more
appropriate tools to plan interventions, design preventa-
tive measures, and forecast the needs of survivor popula-
tions. A latent variable model (LVM) – designed to
Corresponding author:
julianek@umich.edu
Journal of Peace Research
2020, Vol. 57(6) 728–739
ªThe Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0022343320961147
journals.sagepub.com/home/jpr
To continue reading
Request your trial