Innovations in concepts and measurement for the study of peace and conflict
| Author | James Lo,Christopher J Fariss |
| Published date | 01 November 2020 |
| Date | 01 November 2020 |
| DOI | 10.1177/0022343320975200 |
| Subject Matter | Introduction |
Introduction
Innovations in concepts and measurement
for the study of peace and conflict
Christopher J Fariss
Department of Political Science, University of Michigan
James Lo
Department of Political Science and International Relations, University of Southern California
Abstract
The observation, measurement, and analysis of violent and contentious processes are essential parts of the scientific
study of peace and conflict. However, concepts such as the level of repression, the number of individuals killed
during a civil war, or the perception of members of an out-group, are often by definition difficult to observe directly.
This is because governments, non-state groups, NGOs, international organizations, monitoring organizations, and
other actors are not incentivized to make information about their actions systematically observable to analysts. In this
context, latent variable models can play a valuable role by aggregating various behavioral indicators and signals
together to help measure latent concepts of interest that would not otherwise be directly observable. Each of the
articles in this special issue uses some form of a latent variable model or related measurement model to bring together
observable pieces of information and estimate a set of values for the underlying theoretical concept of interest. Each of
the articles pays special attention to the processes that make the observation of peace and conflict processes so
challenging. As we highlight throughout this introductory article, the unifying framework we present in this special
issue is validation. Though the substantive content of each of the articles in this special issue varies, they represent the
diversity of substantive interests that span the study of peace and conflict, broadly conceived. Overall, we hope that
the special issue becomes a standard reference for scholars interested in developing and validating new measurement
models for the study of peace and conflict.
Keywords
measurement models, peace and conflict, scientific process, theoretical concepts
Introduction to the special issue
The observation, measurement, and analysis of violent
and contentious processes, such as the level of repression,
the number of individuals killed during a civil war, or the
perception of members of an out-group, are essential
parts of the scientific study of peace and conflict. How-
ever, governments, non-state groups, NGOs, interna-
tional organizations, monitoring organizations, and
other actors are not always incentivized to make infor-
mation about their actions systematically observable to
analysts. Lack of access and resource constraints on mon-
itoring resources make analyzing contentious political
process all the more difficult for scholars. These issues
are well known to scholars of peace and conflict, who
have collectively devoted considerable resources to the
creation and curation of datasets that are transparently
produced. However, issues of bias and measurement
error are ever-present challenges for this research com-
munity. Fortunately, latent variable models are capable
of integrating and analyzing diverse sources of informa-
tion together and help address issues of bias and error
that arise across different political contexts. In this special
issue, a group of peace and conflict scholars present
Corresponding author:
cjf0006@gmail.com
Journal of Peace Research
2020, Vol. 57(6) 669–678
ªThe Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0022343320975200
journals.sagepub.com/home/jpr
To continue reading
Request your trial