Measures of Foreign Policy Similarity (FPSIM) dataset

The similarity of states’ foreign policy positions is a standard variable in the quantitative, dyadic analysis of international relations. Recent studies routinely rely on Signorino and Ritter’s (1999) S to assess the similarity of foreign policy ties. However, S neglects two fundamental characteristics of the international state system: foreign policy ties are relatively rare and individual states differ in their innate propensity to form such ties. Two chance-corrected agreement indices, Scott’s (1955) π and Cohen’s (1960) κ, represent viable alternatives. Both indices adjust the dyadic similarity score for a large number of common absent ties. Cohen’s κ also takes into account differences in individual dyad members’ total number of ties. The resulting similarity scores have stronger face validity than S. A comparison of their empirical distributions and several replication studies demonstrate that the different types of measures are not substitutable. 

Description and comparison of measures (please cite when using the dataset)

Dataset Version 2.0 (7 December 2017)

Version 2.0 of the FPSIM dataset of measures of foreign policy similarity is now available for download download from my Dataverse.

The dataset provides measures of foreign policy similarity of dyads based on alliance ties (Correlates of War, version 4.1) and UN General Assembly voting (Voeten, version 17) for all members of the international system (Correlates of War system membership list, version 5.0). The alliance data cover the time period from 1816 to 2012, and the UN voting data from 1946 to 2015. The similarity measures include various versions of Ritter and Signorino's S (weighted/non-weighted by material capabilities; squared/absolute distance metrics) as well as the chance-corrected measures Cohen's Kappa and Scott's Pi. The measures based on alliance data come in two versions: one is based on valued alliance ties and the other is based on binary alliance ties.