World Event Interaction Survey Research Articles

Taking It to the Extreme: The Effect of Coalition Cabinets on Foreign Policy

Institutional constraints have been offered by some scholars as an explanation for why multiparty coalitions should be more peaceful than single-party cabinets. Yet others see the same institutional setting as a prescription for more aggressive behavior. Recent research has investigated these conflicting expectations, but with mixed results. We examine the theoretical bases for these alternative expectations about the effects of coalition politics on foreign policy. We find that previous research is limited theoretically by confounding institutional effects with policy positions, and empirically by analyzing only international conflict data. We address these limitations by examining cases of foreign policy behavior using the World Event/Interaction Survey (WEIS) dataset. Consistent with our observation that institutional constraints have been confounded with policy positions, we find that coalitions are neither more aggressive nor more peaceful, but do engage in more extreme foreign policy behaviors. These findings are discussed with regard to various perspectives on the role of institutions in shaping foreign policy behavior.

Unity in Action: Explaining Alignment Behavior in the Middle East

Stephen Walt’s The Origin of Alliances is the most comprehensive, theoretically informed, and widely cited study to date of Middle East alignment behavior during and after the turbulent ‘Arab cold war’ period of the 1950s and 1960s. But Walt may have overstated the discordant character of inter-Arab behavior when he concludes that Arab states ‘balance’ more than they ‘bandwagon’. This study argues that the strength, duration, and ultimate impermanence of regional alignments against Israel indicate that they had a more important basis in ideology than Walt presumes or that realists can accept. Through a graphical portrayal of the results of a factor analysis performed on COPDAB (Conflict and Peace Data Bank) data for the 1948–78 period, and WEIS (World Event Interaction Survey) data for the 1966–93 period, this study establishes that, into the 1970s, the divisions implied by Walt never fully overshadowed the unity in opposition to Israel that was found among parties on the front lines of that confrontation. More specifically, it finds that concordance rather than division prevailed in relations among Israel’s principal adversaries, that concordance had an important basis in conflict with Israel, and that, by the late 1970s, regional politics were far less orderly. It supports a constructivist perspective by concluding that, as a focal point of Arab nationalism, the conflict with Israel imposed normative constraints upon the behavior of Arab leaders.

Emergent Peacemakers: Cataloguing New Patterns of Activity in Post-Cold War Conflict

This paper develops a model using content analysis of chronological event texts to measure third party activity and their impact on outcomes from three case studies of international conflict: The 1990-91 Gulf War; The 1998 war in Kosovo; and the 2002 war in Afghanistan. Data was generated from each of the three conflicts and coded using the World Event Interaction Survey (WEIS). The generated data is used to catalogue participatory levels in post-Cold War conflicts by actor, actor type and level of activity. Data is then compared across cases, each of the three cases showing an increase in the number and influence of third party actors during international conflict and post conflict recovery period.

Cluster-Based Early Warning Indicators for Political Change in the Contemporary Levant

We use cluster analysis to develop a model of political change in the Levant as reflected in the World Event Interaction Survey coded event data generated from Reuters between 1979 and 1998. A new statistical algorithm that uses the correlation between dyadic behaviors at two times identifies clusters of political activity. The transition to a new cluster occurs when a point is closer in distance to subsequent points than to preceding ones. These clusters begin to “stretch” before breaking apart, which serves as an early warning indicator. The clusters correspond well with phases of political behavior identified a priori. A Monte Carlo analysis shows that the clustering and early warning measures are not random; they perform very differently in simulated data sets with similar statistical characteristics. Our study demonstrates that the statistical analysis of newswire reports can yield systematic early warning indicators, and it provides empirical support for the theoretical concept of distinct behavioral phases in political activity.

Measuring international conflict: Developing cross‐country time‐series data

This paper develops a new time series of conflict data from 1948 to the present through statistical estimation of a pooled time‐series of the two most frequently used dyadic events data sets: Conflict and Cooperation Data Bank (COPDAB) (1948–1978) and World Events Interaction Survey (WEIS) (1966‐present). The resulting data series is the longest possible time‐series annual data of dyadic interaction based on daily events. The paper also provides an example of performing a Vector Autoregression (VAR) analysis of conflict and trade in a three‐country setting, utilizing the integrated data. Compared to the COPDAB and WEIS data, the integrated data have a large degree of variation and produce forecasting results that are more complex than those from the COPDAB and WEIS separately.

Behavioral Issues of Rationality in International Interaction

The behavioral pointers of rationality discussed by Langlois and Langlois are used to scan the World Events Interaction Survey data on U.S.-China and U.S.-Soviet relations for evidence of rational behavior in the interaction between states. The null hypothesis that the interaction between states is not compatible with subgame perfect equilibrium play in all cases examined is rejected: states react to each other's past moves; if deviations from equilibrium play are significant in explaining player moves, they determine positive inducement effects; and implied payoff structures reveal a need for strategic management of the relationship. With payoff structures estimated to be prisoner's dilemmas for all players, both bilateral relations begin with the widest tolerance of the other's behavioral choice and the adoption of countervailing strategies. Over time, these nations' strategic interaction evolves toward a sharpening of the bounds they envision for their relationships.

Bilateral Trade and Political Conflict/Cooperation: Do Goods Matter?

This article investigates Granger causality between political conflict/cooperation and bilateral trade. The measures of conflict/cooperation are constructed by accumulating daily events and splicing the two datasets of the Conflict and Peace Data Bank and the World Events Interaction Survey. Trade data from the United Nations include ten commodity groups as well as total trade. Quarterly data are analyzed from the very early 1970s to the early 1990s for four dyads of USA-USSR, USA-China, Turkey-Greece, and Egypt-Israel. Yearly data are investigated from the early 1960s to the early 1990s for 16 dyads. Granger causality between bilateral trade and conflict/cooperation is generally reciprocal in most goods and dyad dependent, but independent of whether or not two countries are political rivals. For USA-USSR and USA-China, however, there is a tendency for bilateral trade to increase in some goods when political relations improve. For USA-USSR, in particular, causality from conflict to trade is pronounced in more goods than causality from trade to conflict. While the effect of cooperation in these dyads is mostly positive, the effect of an increase of trade on conflict is generally ambiguous. For 20 dyads collectively, conflict/cooperation tends to Granger-cause bilateral trade in minerals, iron and steel, fuels, basic manufactures and control and scientific equipment; whereas bilateral trade somewhat more frequently Granger-causes conflict/cooperation in food and live animals, beverages and tobacco, and machines and transport equipment. The concept of strategic goods, much debated in the literature, is further discussed in light of these results. The general result of reciprocal Granger causality calls for a model in which both bilateral trade and conflict/cooperation are simultaneously determined. Such a simultaneous equations model is briefly sketched.

International Conflict and Cooperation: Splicing COPDAB and WEIS Series

Events data sets are used extensively in the study of international relations. The Conflict and Peace Data Bank (COPDAB) data from 1948 to 1978 and the World Events Interaction Survey (WEIS) data since 1966 are the two events data sets that are most widely used. In this article a simple method is proposed to combine (or splice) COPDAB and WEIS events time series in order to create a single spliced time series from 1948 to 1993. Past research on the compatibility of COPDAB and WEIS is inconclusive: some authors claim that COPDAB and WEIS are compatible, while others claim they are not. We here investigate six dyads and show that the compatibility of WEIS and COPDAB depends on the particular dyad in question, but the two series are generally compatible so that COPDAB and WEIS can be spliced into one time series in five out of six dyads. (One dyad contains too many periods of missing reports in the data.) The article is organized in the following sections: The COPDAB and WEIS data sets are described in the first section. The second section reviews the literature about the COPDAB and WEIS compatibility. The third section outlines our approach to the question of compatibility. Statistical tests for compatibility and the method for splicing the two data sets are provided in the fourth section. A summary of our findings is provided in the last section.

Measuring Interstate Affect

This article advances a general formula that can be applied to classified behavioral data to yield a theoretically informed summary measure of interstate affect. Although this measurement procedure is constructed with exclusive reference to expressed affect-that is, a government's feelings of friendliness or hostility toward some external object or actorit is easily adapted to other properties embracing positive, neutral, and negative categories. Once the general measurement model is obtained, certain modifications are introduced that take account of varying time order effects in the development of affective attitudes and the occasional need for monadic-level measurement. The study concludes with an empirical example in which the affect measure is applied to World Event Interaction Survey (WEIS)encoded events exchanged between the United States and the Soviet Union. An actionreaction model fit to bilateral, quarterly affect scores revealed that these two nations follow virtually identical processes in generating affective evaluations of one another.

Controlling instability and conflict through arms transfers: Testing a policy assumption

This paper examines the policy assumption that conventional arms transfers and the conflict behavior of 87 Third World countries are linked causally. It uses World Event Interaction Survey (WEIS) data and U.S. Arms Control and Disarmament Agency data from 1967 through 1976 to measure conflict and arms transfers. Based on a series of regression analyses, the paper's principal conclusions are, in general, that conflict and arms transfers are not linked causally, that variations in the two variables occur simultaneously, and that a deeper understanding of Third World conflict processes is required before those processes can be controlled by reducing conventional arms transfers.

A Comparative Study of the WEIS and COPDAB Data Sets

Data from the Conflict and Peace Data Bank (COPDAB) and the World Event Interaction Survey (WEIS) reflecting United States and USSR cooperation and conflict are examined comparatively. Using the frequencies of events aggregated by month and by year, the portrayal of the history of US-Soviet interaction is first examined visually with graphs presenting the two versions of the activities of the two countries. The graphs show a number of differences in the characterization of changes in levels of cooperation and conflict, with the two data sets disagreeing on the direction of change 29% of the time. Correlations of the pairs of variables from the two data sets indicate a high degree of ‘non-consistency’ for three of the four variables examined. When a series of six hypotheses are tested, four of the six are significantly different when compared across the data sets. When conflict/cooperation ratios are calculated as a measure of ‘tension’ between the two states, these also differ considerably for WEIS and COPDAB. Whether the comparison of the COPDAB and WEIS versions of US-Soviet interaction is in the form of graphs, correlations of pairs, hypothesis tests on principles of interaction, or correlations of tension ratios, the two data sets seldom match up well. Overall, the choice of one data set rather than the other is likely to introduce variation in outcomes relating to theoretical principles and policy plans.

Let the User Beware

Two reports in this issue converge with findings which, in essence, constitute a warning to researchers. Professors Jack Vincent and Llewellyn Howell have agreed that it is unwise to place great trust in the World Event Interaction Survey (WEIS) and the Conflict and Peace Data Bank (COPDAB). Both scholars have carried out series of data analyses, using first one event data collection and then the other, and both have found that the results are importantly different. Neither Howell nor Vincent are ready, at this time, to recommend the use of one data set and the avoidance of the other. Both make references to probable practices of 'over-collecting' and 'under-collecting', but their investigations to date have not provided detail sufficient to pinpoint evidences of faulty workmanship in the data acquisition process. Equally for foreign policy making and international theory, Vincent and Howell have sounded an alarm with regard to the probable validity and reliability of the two major sets of international event data. They are not the first to suspect event data, however. Objections have been voiced and doubts raised, with persistence and vigor, from the first appearance of event studies. Almost any group of critics can be relied upon to react with 'standard' queries, among them: (1) Won't event files, being based on daily public reporting of world news, be biased seriously because of the selection and suppression of stories by reporters, editors, and owners of the mass media?; (2) Since heads of state, foreign ministers, intelligence chiefs, and military leaders do much of their work in foreign affairs in secret, how can event data collections be regarded as valid-how can they reflect the real circumstances of world affairs, in the face of so much secret activity?;

Cooperation and Conflict in Foreign Policy Behavior: Reaction and Memory

This study examines the dynamics of international cooperation and conflict in the foreign policy of a selected number of nations during the contemporary era. Two specific types of conceptions are developed and examined empirically. The first of these is based on the idea that conflict and cooperation behaviors are involved in separate dynamics, based upon long- and short-term institutional memory and upon reactivity. The second broad model examines hypotheses that cooperation and conflict interact with one another as the evolve over time. The first model, the behavior-begets-behavior conception, is more substantially supported by empirical evidence than is the mix of behaviors-begets-mix of behaviors model. In large part, these results are upheld across two major collections of event interaction data, the World Event Interaction Survey (WEIS) and the Conflict and Peace Data Bank (COPDAB). Thus, it appears that adequate understanding of contemporary foreign policy should be based upon the high degree of reactivity which exists among contemporary nation-states.