Analyzing the Dynamics of International Mediation Processes

University Of Kansas Center For Research Inc, Lawrence KS

Investigators

Abstract

This project examines the dynamics of third-party international mediation using statistical time-series analyses of political event data. Third-party mediation is one of the most common international responses to political conflict. Studies show that the use of mediation has increased following the end of the Cold War. Surprisingly, there are few systematic studies of mediation. Those that do exist focus generally on relatively static contextual factors, such as the attributes of the conflict and the prior relationship between the mediator and protagonists, rather than on dynamic factors - both contextual and process - that may contribute to the success or failure of mediation activities. In contrast, the extensive qualitative literature provides numerous hypotheses about dynamic aspects of mediation. These studies, however, are primarily case studies, often are undertaken by mediation practitioners, exhibit little cumulation and, when taken as a whole, are rife with contradictory assertions. This research formally tests a number of the hypotheses embedded in the theoretical and qualitative literatures on mediation, suing automated coding of event data from news-wire sources and employing time-series and event-history methods available in standard statistical packages. Event data provide a rich set of indicators about the results of mediation, the political circumstances of the mediation (for example, prior military success or failure by the protagonists), and the various strategies employed by the mediating parties). The research looks at the factors that influence (1) whether mediation is accepted by the parties in a conflict, (2) whether formal agreements are reached, and (3) whether the agreements actually reduce the level of conflict. The project focuses initially on conflicts in the Middle East, a region where the principal investigators have substantial field experience. The TABARI machine-coding system - an open-source successor to the NSF-supported KEDS coding program - is adapted to code events according to the new IDEA coding framework developed by Taylor, Jenkins and Bond. The broader research community has access to the revised coding program, coding dictionaries, and datasets via a web site. After refining the statistical tests on the Middle East case, the analysis is extended to event data on conflicts in the former Yugoslavia and West Africa. Finally, a subset of tests is applied to the historical crises in Leng's Behavioral Correlates of War dataset and to the contemporary crises being documented in Alker's Conflict Early Warning System project. The research uses three types of statistical models. The first set of models employs interval-level methods on scaled and aggregated event data, using methods similar to those found in econometric analysis. The sensitivity of the event scaling system also is examined. Second, mediation is explored using the "time series cross-sectional logit" technique that is used widely in the "democratic peace" research and other war-initiation literature. This work takes into consideration modifications suggested by Beck, Katz, King and others to improve the effectiveness of this methodology. Finally, event-history and other duration models are estimated; this approach should be particularly useful given the large random element in the timing of many conflicts. This study contributes in both the academic and practical arenas. It provides an extensive empirical analysis of hypotheses about the dynamics of the mediation process. Those empirical generationalizations can inform the efforts of individuals engaged in actual international mediation. The project also continues the development of research tools and datasets that can be used in a variety of statistical studies of international behavior.