Cornwall

Working Papers

  • 2019. Civil war outcome, democratization, and ethnic support [Abstract]
       SocArXiv [Download]
    • Abstract
      What are the conditions under which post-conflict elections take place and why do only few post-conflict elections result in democratic post-conflict orders? This is the main puzzle raised in this paper. The argument is that the decision to stage elections and democratize is highly strategic and depends on the ethnic size of the post-conflict government leader. Especially post- conflict leaders with large ethnic support are able to utilize quick post-conflict elections to stabilize and legitimize their political position. However, in the long run the government leaders prefer institutions that maximize their time in office and therefore will hinder full scale democratization. The empirical analysis demonstrates that the timing of elections and democratization depend on the ethnic support of the new government and the civil war outcome.
    • Manuscript in PDF
  • 2019. Predicting the severity of civil wars [Abstract]
        An actor-centric approach
        (with Gökhan Çiflikli and Altaf Ali)
       SocArXiv [Download]
    • Abstract
      This paper introduces an actor-centric approach to predict the severity of conflict. We argue that the prediction of conflict severity needs to focus on the actors that are responsible for conducting armed violence. Hence, we predict the severity of conflict in government-rebel organization dyads. We introduce predictors that focus especially on rebel organization characteristics, behavior, and the conflict network they are are embedded in. Our statistical learning approach relies on random forests to predict the severity of conflict. We demonstrate that our model is performs especially well distinguishing high levels of severity from very low levels.
    • Manuscript in PDF
  • 2019. We predict conflict better than we thought! [Abstract]
        Taking time seriously when evaluating predictions in Binary-Time-Series-Cross-Section-Data
        (with Gökhan Çiflikli)
       SocArXiv [Download]
    • Abstract
      Efforts to predict civil war onset, its duration, and subsequent peace have dramatically increased. Nonetheless, by standard classification metrics the discipline seems to make little progress. Some remedy is promised by particular cross-validation strategies and machine learning tools, which increase accuracy rates substantively. However, in this research note we provide convincing evidence that the predictive performance of conflict models has been much better than previously assessed. We demonstrate that standard classification metrics for binary outcome data are prone to underestimate model performance in a binary-time-series-cross-section context. We argue for temporal residual based metrics to evaluate cross-validation efforts in binary-time-series-cross-section and test these in Monte Carlo experiments and existing empirical studies.
    • Manuscript in PDF
  • 2019. Strategic Rebel Networks: Fighting durations in a spatial duration framework [Abstract]
        (with Julian Wucherpfennig)
       SocArXiv [Download]
    • Abstract
      Recent research on multi-actor civil wars highlights that rebel organizations condition their conflict behavior on that of other rebel organizations, with competition and free-riding constituting the core theoretical mechanisms. We provide a new actor-centric approach to explicitly model strategic interdependence in multi-actor civil wars. We argue that competitive dynamics dominate strategic behavior between rebel organizations, but these can be offset by incentives to free-ride in cases where the underlying incompatibility displays public good characteristics. Based on a network game theoretic model, we derive a statistical framework that allows for a direct test of strategic interdependence. We find that the estimated duration interdependence is positive, that is weaker in secessionist conflicts, and that modeling this interdependence explicitly outweighs existing empirical measures of interdependence (e.g. number of organizations). Finally, we demonstrate that the model fit of rebel organizations' fighting durations can be improved by taking strategic interdependence into account.
    • Manuscript in PDF

Journal Articles

  • 2018. Multiple Imputation Using Gaussian Copulas [Abstract]
        (with Florian M. Hollenbach, Iavor Bojinov, Shahryar Minhas, Michael D. Ward, and Alexander Volfovsky)
       Sociological Methods and Research. Forthcoming.
    • Abstract
      Missing observations are pervasive throughout observational research, especially in the social sciences. Despite multiple approaches to dealing adequately with missing data, many scholars still rely on list-wise deletion. In this article, we present a simple to use approach to multiple imputation. We show that using Gaussian copulas for multiple imputation allows scholars to attain estimation results that have good coverage and small bias. Using simulated as well as observational data from published social science research we compare imputation via Gaussian copulas with two other widely used imputation methods: MICE and Amelia II. The three approaches perform relatively similarly. Importantly, however, imputation via the Gaussian copula is simple and does not require the researcher to undertake any transformation of the data or specification of distributional assumptions for individual variables but returns a valid posterior density of the imputed data.
    • Manuscript in PDF
  • 2018. Born weak, growing strong: [Abstract]
        Anti-government protests as a signal of rebel strength in the context of civil wars.
        (with Bahar Leventoglu)
       American Journal of Political Science. Vol.62, No.3, pp. 581-596.
    • Abstract
      All rebel organizations start weak, but how do they grow and achieve favorable conflict outcomes? We present a theoretical model that allows for rebel organizations to gain support beyond their `core' and build their bargaining power during fighting. We highlight that rebel organizations need to win over crucial parts of society to generate the necessary support that allows them to attain favorable civil conflict outcomes. We find empirical support for the argument that low income individuals who initially fight the government (rebel organizations) have to convince middle-class individuals to turn out against the government to gain government concessions. Empirically, we demonstrate that government concessions in the form of peace agreements and the onset of negotiations become more likely when protest occurs in the context of civil conflicts.
    • Manuscript in PDF
    • Replication Materials
  • 2017. Splitting It Up: [Abstract]
        The spduration Split-Population Duration Regression Package for Time-varying Covariates.
        (with Andreas Beger, Daniel W. Hill, Shahryar Minhas and Michael D. Ward)
       The R Journal. Vol.9, No.2, pp. 474-486.
    • Abstract
      We present an implementation of split-population duration regression in the spduration package for R that allows for time-varying covariates. The statistical model accounts for units that are immune to a certain outcome and not part of the duration process the researcher is primarily interested in. We provide insights that if immune units exist, we can significantly increase the predictive performance compared to standard duration models. The package includes estimation and several post-estimation methods for split-population Weibull and Loglogistic models. We provide an empirical application to data on military coups.
    • Manuscript in PDF
    • R Package
  • 2017. Forecasting in peace research. [Abstract]
        Introduction to Special Issue on Forecasting
        (with Håvard Hegre, Håvard Mokleiv Nygård and Julian Wucherpfennig)
       Journal of Peace Research. Vol.54, No.2, pp. 113-124.
    • Abstract
      Prediction and forecasting have now fully reached peace and conflict research. We define forecasting as predictions about unrealized outcomes given model estimates from realized data, and predictions more generally as the assignment of probability distributions to realized or unrealized outcomes. Increasingly, scholars present within- and out-of-sample prediction results in their publications and sometimes even forecasts for unrealized, future outcomes. The articles in this special issue demonstrate the ability of current approaches to forecast events of interest and contributes to the formulation of best practices for forecasting within peace research. We highlight the role of forecasting for theory evaluation and as a bridge between academics and policymakers, summarize the contributions in the special issue, and provide some thoughts on how research on forecasting in peace research should proceed. We suggest some best practices, noting the importance of theory development, interpretability of models, replicability of results, and data collection.
  • 2016. Firewall? Or Wall on Fire? [Abstract]
        A unified framework of conflict contagion and the role of ethnic exclusion
        (with Shahryar Minhas and Michael D. Ward)
       Journal of Conflict Resolution. Vol.61, No.6, pp. 1151-1173.
    • Abstract
      While some borders are real firewalls against conflicts others appear like tinder just waiting for the smallest spark. Only recently has research focused on the transnational perspective of conflict and current research has focused mostly on isolated aspects of this phenomenon. In this article, we provide a unified framework for conflict contagion that takes into account receiver, sender, dyad, and network effects. This is a novel perspective on conflict contagion and our empirical results suggest that distinguishing between sender and receiver effects allows for a better understanding of spill-over effects. We provide insights that especially ethnic excluded groups impact on the risk of countries sending and receiving conflicts from its neighbors.
  • 2016. The Flight of White-Collars: [Abstract]
        Civil conflict, availability of medical service providers, and public health.
        (with Arzu Kibris)
       Social Science and Medicine. Vol.149, pp. 93-103.
    • Abstract
      Civil conflicts devastate public health both in the short run and in the long run. Analyzing novel data sets that include yearly information on public health and the availability of health professionals across provinces in Turkey in the 1964–2010 period, we provide empirical evidence for our theoretical argument that a major mechanism through which civil conflicts exert their long term negative influences on public health is by discouraging medical personnel to practice in conflict regions. We also assess the effectiveness of certain policy measures that Turkish governments have tried out over the years to counteract this mechanism. Our results reveal that the long running civil conflict in Turkey has been driving away doctors and other highly trained medical personnel from conflict areas and that mandatory service requirements do help counteract this flight.
  • 2015. Every Story Has a Beginning, Middle, and an End (But Not Always in That Order): [Abstract]
        Predicting duration dynamics in a unified framework
        (with Daina Chiba and Michael Ward)
       Political Science Research and Methods, Vol.3 No.3, pp. 515-541.
    • Abstract
      There are three fundamental duration dynamics of civil conflicts: Time until conflict onset, conflict duration, and time until conflict recurrence. Theoretical and empirical models of war usually focus on one or at most two aspects of these three important duration dynamics. We argue that duration forecasting of conflict needs to incorporate all three conflict dynamics as unobserved factors might impact on all three of these dynamics. We present a new split-population seemingly-unrelated duration estimator that treats pre-conflict duration, conflict duration, and post-conflict duration as interdependent processes thus permitting improved predictions about the onset, duration, and recurrence of civil conflict. Our findings provide support for the more fundamental idea that prediction is dependent on a good approximation of the theoretically implied underlying data generating process. In addition, we account for the fact that some countries might never experience these duration dynamics or become immune after experiencing them in the past.
  • 2014. Data and Progress in Peace and Conflict Research. [Abstract]
        (with Kristian Skrede Gleditsch and Andrea Ruggeri)
       Journal of Peace Research, Vol.51, No.2, pp. 301-314.
    • Abstract
      We highlight how efforts to collect systematic data on conflict have helped foster progress in peace and conflict research. The Journal of Peace Research has played a key role in these developments, and has become a leading outlet for the new wave of disaggregated conflict data. We survey progress in the development of conflict data and how this interacts with theory development and progress in research, drawing specifically on examples from the move towards a greater focus on disaggregation and agency in conflict research. We focus on disaggregation in three specific dimensions, namely the resolution of conflict data, agency in conflict data, and the specific strategies used in conflict, and we also discuss new efforts to study conflict processes beyond the use of violence. We look ahead to new challenges in conflict research and how data developments and the emergence of ‘big data’ push us to think harder about types of conflict, agency, and the ‘right’ level of aggregation for querying data and evaluating specific theories.
  • 2013. Anti-government Networks in Civil Conflicts. [Abstract]
        How Network Structures Affect Anti-government Behavior.
       (with Cassy Dorff, Max Gallup, Simon Weschle, and Michael D. Ward)
       American Journal of Political Science, Vol.57, No.4, pp. 892-911.
    • Abstract
      In this article, we combine a game-theoretic treatment of public goods provision in networks with a statistical network analysis to show that fragmented opposition network structures lead to an increase in conflictual actions. Current literature concentrates on the dyadic relationship between the government and potential challengers. We shift the focus toward exploring how network structures affect the strategic behavior of political actors. We derive and examine testable hypotheses and use latent space analysis to infer actors’ positions vis-à-vis each other in the network. Network structure is examined and used to test our hypotheses with data on conflicts in Thailand from 2001 to 2010. We show the influential role of network structure in generating conflictual behavior.
  • 2013. Stepping into the Future: [Abstract]
        A new generation of conflict forecasting models.
       (with Michael D. Ward, Cassy Dorff, Max Gallup, Florian M. Hollenbach, Anna Schultz, and Simon Weschle)
       International Studies Review, Vol.15, No.4, pp. 473-490.
    • Abstract
      Developing political forecasting models not only increases the ability of political scientists to inform public policy decisions, but is also relevant for scientific advancement. This article argues for and demonstrates the utility of creating forecasting models for predicting political conflicts in a diverse range of country settings. Apart from the benefit of making actual predictions, we argue that predictive heuristics are one gold standard of model development in the field of conflict studies. As such, they shed light on an array of important components of the political science literature on conflict dynamics. We develop and present conflict predictions that have been highly accurate for past and subsequent events, exhibiting few false-negative and false-positive categorizations. Our predictions are made at the monthly level for 6-month periods into the future, taking into account the social–spatial context of each individual country. The model has a high degree of accuracy in reproducing historical data measured monthly over the past 10 years and has approximately equal accuracy in making forecasts. Thus, forecasting in political science is increasingly accurate. At the same time, by providing a gold standard that separates model construction from model evaluation, we can defeat observational research designs and use true prediction as a way to evaluate theories. We suggest that progress in the modeling of conflict research depends on the use of prediction as a gold standard of heuristic evaluation.
  • 2012. Ethnicity, the State, and the Duration of Civil Wars. [Abstract]
       (with Julian Wucherpfennig, Kristian Skrede Gleditsch, and Lars-Erik Cederman)
       World Politics, Vol.64, No.1, pp. 79-115.
    • Abstract
      Previous research has focused primarily on how ethnicity may trigger civil war, and its effect on conflict duration remains disputed. Rather than treating conflict as a direct consequence of ethnic cleavages, the authors argue that ethnicity per se does not affect civil war duration. Instead, its effect depends on its relationship to political institutions. They employ a dyadic approach that emphasizes the political context in which both government leaders and nonstate challengers can capitalize on the ascriptive nature of ethnicity. They show that although states can initially benefit from politicizing ethnic relations, once violent conflict breaks out, such policies may backfire on the government and make it difficult for incumbent governments to accept settlements that could terminate conflicts. Past policies of ethnic exclusion also benefit rebel organizations fighting the government, since the resulting grievances increase collective group solidarity and render individual fighters more cost tolerant. Using a new data set that codes the nexus between rebel organizations and ethnic groups, as well as information on ethnopolitical exclusion, the authors find considerable support for their propositions.
  • 2011. Expecting Elections: Interventions, Ethnic Support, and the Duration of Civil Wars [Abstract]
       Journal of Conflict Resolution, Vol.55, No.6, pp. 909-937.
     
    • Abstract
      International organizations (IOs) frequently link their military interventions with democratization efforts in the target state. However, existing research suggests that these attempts often fail. This article analyzes the conditions under which interventions by IOs shorten or prolong civil war dyads. When militarily strong rebel groups with low public support expect externally enforced democratization, they have incentives to continue fighting. These incentives arise when democratization leads to power shifts that cause commitment problems for belligerents with high popular support. Cox hazards models are used to test the article’s hypotheses on a new data set on African rebel leaders’ ethnicity. The results demonstrate that IO interventions with democratization mandates are only associated with shorter conflicts if rebel leaders come from ethnic groups representing more than 10 percent of a country’s population. IO interventions without democratization mandates are not associated with shorter conflict duration and show no interaction effect with the rebels’ ethnic support.

Funded Projects

  • Predicting the escalation of conflict (Economic and Social Research Council)
    • The key objective of this research project is to forecast conflict escalation of intra-state con- flicts. In the past few years an increasing number of researchers are becoming interested in forecasting conflicts. This goes along with a more general interest in political science forecasting of elections, regime transitions, and terrorism. Methodological, technological, and data-related (big data) advances are making it possible to forecast one of the most complicated phenomena: social behavior. Existing forecast models of conflict have very much focused on the occurrence of conflict but not their intensity. However, forecasting the intensity of conflicts is important to implement adequate policy responses, build resilience, and prepare early action. Especially, in the context of limited resources, when reacting to conflict around the world it is not only important to know if conflict occurs, but when and where it escalates. This research projects develops split-population forecasting models that predict both, the occurence and escalation of conflict. The research has clearly impact oriented strategy by delivering software packages, web-based prediction tools, and implementation strategies for government, non-government, and corporate users.
  • Who joins and who fights? (Gerda Henkel Stiftung)
    • In civil conflicts, rebel groups, ethnic groups, and other actors form coalitions while challenging the government. Such alliances frequently break up and armed groups fight each other. This proposal highlights that existing research on civil conflicts has limited insights to when al- liances form and when they disintegrate. We propose a network perspective on collaboration and competition between violent ethnic groups that enables predictions about when groups join alliances and when infighting occurs.


Cornwall


Welcome

I am currently Associate Professor at University College London in the Department of Political Science. My research focuses on the strategic nature of civil conflicts and the prediction of its dynamics. Previously, I was a Post-doc at Duke University in the Department of Political Science. I received my PhD in Government from the University of Essex. My work has been awarded the Stuart Bremer Award (Peace Science Assocation), the Dina Zinnes Award (International Studies Association), and the Best Paper Award in the Conflict Processes Section (American Political Science Association).

News

  • August 2018: APSA paper: Civil war outcome, democratization, and ethnic support. [Paper]
  • August 2018: APSA paper with Janina Beiser: Ethnic coalitions and the logic of political survival in authoritarian regimes. [Paper]
  • April 2018: ISA paper: We are better at predicting conflict than we thought. [Paper]
  • August 2017: BlogPost: Being Policy Relevant in Peace Research means Forecasting. [Read]

  • Site Design