This article summarizes the Correlates of War Intergovernmental Organizations (IGO) Version 3.0 datasets. The new datasets include information about the population of IGOs in the international system and state participation in those formal international institutions from 1816 to 2014. Consistent with Versions 2.0 and 2.3, Version 3.0 of the IGO data comes in three forms: country-year, IGO-year, and joint dyadic membership. This article briefly describes the data collection process and identifies important changes to the dataset before moving to analyze fundamental patterns in the data. Most notable among the changes from earlier versions of the data is the inclusion of annual membership data for the 1815–1964 time period. In addition, we present information about the overall trends in the institutionalization of cooperation at both the global and regional levels, with the latter focusing on the interesting membership dynamics in Asia and Africa. We then track and discuss patterns in state memberships and examine how these changes manifest in the dyadic data. The article concludes with a discussion of how the COW IGO 3.0 data compare to other prominent datasets on state participation in international institutions and highlights some new areas of research that will benefit from the release of the updated IGO membership dataset.

This paper explores the question of how major powers signal support for their protégés. We develop a theory that explains why major powers show support for some protégés using highly visible “frontstage” signals of support, while supporting other protégés through less visible, but nonetheless costly, “offstage” signals. From an international strategic perspective, it is puzzling that major powers do not always send the most visible signal possible. We argue that this can be explained by considering the domestic environments in which the leaders of major powers and protégés operate. Focusing particularly on the United States as we develop our theory, we argue that the US will prefer to send offstage signals of support for more autocratic protégés for several reasons. First, sending frontstage support signals for autocracies would expose US leaders to charges of hypocrisy. Second, frontstage signals of support for autocracies face an impediment to credibility because of the public backlash in the United States that overt support for dictators could generate. Third, many autocratic protégés would be reluctant to accept a frontstage signal of support from the US because it could undermine their regime stability. We test our theory in a data set that records various support signals sent by the United States for other countries between 1950 and 2008, finding strong support for our expectations. We also find evidence of the causal mechanisms posited by our theory in a case study of relations between the US and the Shah's Iran.

Major powers signal support for protégés in order to reassure them and deter harm against them. Yet, it is not always clear how to identify who a major power's protégés are or the degree of support signaled. Major powers have a variety of complementary signals to choose among, including alliances, arms transfers, joint military exercises, and others. It can be difficult to weigh the importance of individual signals, especially since different major powers do not deploy each signal in the same way. We address this challenge using a Bayesian latent measurement model, which provides a theoretically coherent means of identifying the overall level of support signaled by a major power for a protégé. Our approach yields a cross-sectional time-series dataset, providing a continuous measure of the degree of support signaled by major powers for all minor powers from 1950 to 2012. Our model also provides insights regarding which signals of support are most informative when sent by each major power. We find considerable variation among major powers regarding which of their signals are most meaningful, but in general alliances and military exercises tend to be among the most important signals. In further applications using our latent measure, we also assess under which conditions major powers are likely to increase their signals of support for protégés, as well as predict whether a major power will intervene in conflicts involving its protégés.

No abstract

In a recent article, Gibler, Miller, and Little (2016) (GML) conduct an extensive review of the Militarized Interstate Dispute (MID) data between the years 1816 and 2001, highlighting possible inaccuracies and recommending a substantial number of changes to the data. They contend that, in several instances, analyses with their revised data lead to substantively different inferences. Here, we review GML's MID drop and merge recommendations and reevaluate the substantive impact of their changes. We are in agreement with about 76 percent of the recommended drops and merges. However, we find that some of the purported overturned findings in GML's replications are not due to their data, but rather to the strategies they employ for replication. We reexamine these findings and conclude that the remaining differences in inference stemming from the variations in the MID data are rare and modest in scope.