[W]hat lies beneath: using latent networks to improve spatial predictions of political violence

Max Gallop, Cassy Dorff, Shahryar Minhas

Research output: Contribution to journalArticle

Abstract

Spatial interdependencies commonly drive the spread of violence in civil conflict. To address such interdependence, scholars often use spatial lags to model the diffusion of violence, but this requires an explicit operationalization of the connectivity matrices that represent the spread of conflict. Unfortunately, in many cases, there are multiple competing processes that facilitate the spread of violence making it difficult to identify the true data-generating process. We show how a network driven methodology can allow us to account for the spread of violence, even in the cases where we cannot directly measure the factors that drive diffusion. To do so, we estimate a latent connectivity matrix that captures a variety of possible diffusion patterns. We use this procedure to study intrastate conflict in eight conflict-prone countries and show how our framework enables substantially better predictive performance than canonical spatial lag measures. We also investigate the circumstances under which canonical spatial lags suffice, and those under which a latent network approach is beneficial.
Original languageEnglish
Pages (from-to)1-32
Number of pages32
JournalInternational Studies Quarterly
Publication statusAccepted/In press - 17 Jan 2020

Fingerprint

political violence
violence
operationalization
interdependence
methodology
performance

Keywords

  • spatial analysis
  • network analysis
  • conflict
  • civil war

Cite this

@article{9052836f67b74f6b8755d7723c2faac7,
title = "[W]hat lies beneath: using latent networks to improve spatial predictions of political violence",
abstract = "Spatial interdependencies commonly drive the spread of violence in civil conflict. To address such interdependence, scholars often use spatial lags to model the diffusion of violence, but this requires an explicit operationalization of the connectivity matrices that represent the spread of conflict. Unfortunately, in many cases, there are multiple competing processes that facilitate the spread of violence making it difficult to identify the true data-generating process. We show how a network driven methodology can allow us to account for the spread of violence, even in the cases where we cannot directly measure the factors that drive diffusion. To do so, we estimate a latent connectivity matrix that captures a variety of possible diffusion patterns. We use this procedure to study intrastate conflict in eight conflict-prone countries and show how our framework enables substantially better predictive performance than canonical spatial lag measures. We also investigate the circumstances under which canonical spatial lags suffice, and those under which a latent network approach is beneficial.",
keywords = "spatial analysis, network analysis, conflict, civil war",
author = "Max Gallop and Cassy Dorff and Shahryar Minhas",
note = "This is a pre-copyedited, author-produced version of an article accepted for publication in International Studies Quarterly following peer review. The version of record Gallop, M, Dorff, C & Minhas, S 2020, '[W]hat lies beneath: using latent networks to improve spatial predictions of political violence', International Studies Quarterly, pp. 1-32. is available online at: https://academic.oup.com/isq.",
year = "2020",
month = "1",
day = "17",
language = "English",
pages = "1--32",
journal = "International Studies Quarterly",
issn = "0020-8833",

}

[W]hat lies beneath : using latent networks to improve spatial predictions of political violence. / Gallop, Max; Dorff, Cassy; Minhas, Shahryar.

In: International Studies Quarterly, 17.01.2020, p. 1-32.

Research output: Contribution to journalArticle

TY - JOUR

T1 - [W]hat lies beneath

T2 - using latent networks to improve spatial predictions of political violence

AU - Gallop, Max

AU - Dorff, Cassy

AU - Minhas, Shahryar

N1 - This is a pre-copyedited, author-produced version of an article accepted for publication in International Studies Quarterly following peer review. The version of record Gallop, M, Dorff, C & Minhas, S 2020, '[W]hat lies beneath: using latent networks to improve spatial predictions of political violence', International Studies Quarterly, pp. 1-32. is available online at: https://academic.oup.com/isq.

PY - 2020/1/17

Y1 - 2020/1/17

N2 - Spatial interdependencies commonly drive the spread of violence in civil conflict. To address such interdependence, scholars often use spatial lags to model the diffusion of violence, but this requires an explicit operationalization of the connectivity matrices that represent the spread of conflict. Unfortunately, in many cases, there are multiple competing processes that facilitate the spread of violence making it difficult to identify the true data-generating process. We show how a network driven methodology can allow us to account for the spread of violence, even in the cases where we cannot directly measure the factors that drive diffusion. To do so, we estimate a latent connectivity matrix that captures a variety of possible diffusion patterns. We use this procedure to study intrastate conflict in eight conflict-prone countries and show how our framework enables substantially better predictive performance than canonical spatial lag measures. We also investigate the circumstances under which canonical spatial lags suffice, and those under which a latent network approach is beneficial.

AB - Spatial interdependencies commonly drive the spread of violence in civil conflict. To address such interdependence, scholars often use spatial lags to model the diffusion of violence, but this requires an explicit operationalization of the connectivity matrices that represent the spread of conflict. Unfortunately, in many cases, there are multiple competing processes that facilitate the spread of violence making it difficult to identify the true data-generating process. We show how a network driven methodology can allow us to account for the spread of violence, even in the cases where we cannot directly measure the factors that drive diffusion. To do so, we estimate a latent connectivity matrix that captures a variety of possible diffusion patterns. We use this procedure to study intrastate conflict in eight conflict-prone countries and show how our framework enables substantially better predictive performance than canonical spatial lag measures. We also investigate the circumstances under which canonical spatial lags suffice, and those under which a latent network approach is beneficial.

KW - spatial analysis

KW - network analysis

KW - conflict

KW - civil war

UR - https://academic.oup.com/isq

M3 - Article

SP - 1

EP - 32

JO - International Studies Quarterly

JF - International Studies Quarterly

SN - 0020-8833

ER -