ABSTRACT
More than three decades have passed since the Correlates of War project received its initial grant to collect data on the causes and consequences of war. Since then, a multitude of datasets have been born covering a wide variety of phenomena in international relations. We now have data on alliances, material capabilities, diplomatic recognition, international organizations and their memberships, inter- and intra-state conflict, regime type, changes of government, cultural composition of states, and several forms of political rebellion. Clearly, these datasets have helped to provide a foundation that has allowed the scientific study of war to make remarkable progress in recent years. Nevertheless, many areas within international politics are still arguably theory-rich and data-poor, and, as a result, much of what passes as theory is based largely upon speculation rather than arguments constructed from hard evidence. Despite the best efforts by scholars in the field, the empirically based knowledge on the causes, courses, and consequences of interstate conflict is still far from complete. The literature on interstate conflict is both extensive and fragmentary, at the same time. Several facets have been rigorously pursued such as the war-to-war question, the success of deterrent policies, the conflict propensities of regimes, the interplay between economics and war, and the relationship between system structure and conflict. However, our results are often not comparable due to conceptual and methodological disagreements or different spatial–temporal domains (Bremer, 1993). Meanwhile, other important questions within the conflict puzzle lay dormant because the data needed were unavailable. In contrast to our colleagues in American politics, who have access to assembled data from numerous government sources or opinion surveys, we must collect and process data while being constrained by finite budgets and limited time that ultimately determines what we can and cannot investigate.
